TWI808142B - Methods and systems for controlling tintable windows with cloud detection - Google Patents

Abstract

Methods and systems for controlling tintable windows based on cloud detection.

Description

Method and system for controlling tintable windows with cloud detection

The disclosure generally relates to the configuration of sensing elements for detecting cloud cover conditions, and in detail, relates to an infrared cloud detector system and a method for detecting cloud cover conditions thereof.

Detecting cloud cover can be an important part of the decision to put equipment into operation, for example, at a robotic weather station, since astronomers may want to detect clouds that can interfere with their observations. Conventional methods of mapping the sky to detect cloud cover rely on expensive imaging devices that typically rely on visible light measurements.

Certain aspects relate to a controller for controlling the tint of one or more tintable windows in a zone of a building. The controller includes a computer readable medium having control logic configured to determine a tint level for the zone of one or more tintable windows based on a cloud condition based on one or both of light sensor readings and infrared sensor readings. The controller further includes a processor in communication with the computer readable medium and with a local window controller of the tintable window. The processor is configured to determine the cloud condition based on one or both of light sensor readings and infrared sensor readings, calculate the tint level for the zone of one or more tintable windows based on the determined cloud condition, and send tint commands to a local window controller via a network to convert the tint of the zone of the tintable window to the calculated tint level. In some aspects, the hue level is determined based on a cloud condition that is likely to occur at a future time.

Certain aspects relate to a method of controlling the tint of a zone of one or more tintable windows of a building. The method includes determining a cloud condition based on one or both of light sensor readings and infrared sensor readings, calculating a tint level for the zone of one or more tintable windows based on the determined cloud condition, and communicating a tint command to a local window controller via a network to convert the tint level of the zone of the tintable window to the calculated tint level. Certain aspects relate to methods and systems for controlling tint levels of tintable windows through cloud detection. In some aspects, the hue level is calculated based on cloud conditions that are likely to occur at a future time.

Some aspects relate to infrared cloud detector systems. In some aspects, an infrared cloud detector system includes: an infrared sensor configured to measure a sky temperature based on infrared radiation received within its field of view; an ambient temperature sensor configured to measure an ambient temperature; and logic configured to determine a cloud condition based on a difference between the measured sky temperature and the measured ambient temperature.

In some aspects, an infrared cloud detector system includes: an infrared sensor configured to measure sky temperature based on infrared radiation received within its field of view; an ambient temperature sensor configured to measure an ambient temperature; a light sensor configured to measure an intensity of visible light; and logic configured to determine a cloud condition. If a time of day is between a time before sunrise and a second time after sunrise or between a third time before sunset and sunset, the logic is configured to determine the cloud condition based on a difference between the measured sky temperature and the measured ambient temperature. If the time of day is between the second time after sunrise and the third time before sunset, the logic is configured to determine the cloud condition based on the measured intensity of visible light from the light sensor.

Certain aspects relate to the infrared cloud detector method. In some aspects, an infrared The line cloud detector method includes receiving a sky temperature reading from an infrared sensor and an ambient temperature reading from an ambient temperature sensor, calculating a difference between the sky temperature reading and the ambient temperature reading, and determining a cloud condition based on the calculated difference between the sky temperature reading and the ambient temperature reading.

In some aspects, an infrared cloud detector method includes receiving a sky temperature reading from an infrared sensor, an ambient temperature reading from an ambient temperature sensor, and an intensity reading from a light sensor, and determining whether a time of day is: (i) between a first time before sunrise and a second time after sunrise or between a third time before sunset and sunset; (ii) between the second time after sunrise and a third time before sunset; (iii) after (i) and between (i) ii) before; or (iv) after (iii) and before (i). If the time of day is (i), (iii) or (iv), the cloud condition is determined based on a difference between the measured sky temperature and the measured ambient temperature. If the time of day is (iii), then the cloud condition is determined based on the intensity reading received from the light sensor.

These and other features and embodiments will be described in more detail below with reference to the drawings.

100: Infrared Cloud Detector

101: shell

102: cover

104: porosity or thinned part

106: first surface

108: second surface

110:IR sensor

112: Imaginary axis

114: Conical field of view

130: Ambient temperature sensor

140: Processor

220: The first curve

222: second curve

230: first curve

232: second curve

300:Infrared Cloud Detector System

310: Infrared Cloud Detector

312: shell

314: infrared sensor

315: Conical Field of View

316: Ambient temperature sensor

320: External Visible Light Sensor

330: room

332: Tintable windows

334: table

340: controller

400:Infrared Cloud Detector System

401: Multi-sensor device

410: shell

411: cover

412: Thinned part

414: base part

420: Ambient temperature sensor

440:Visible light sensor

442: Orientation axis

452: The first infrared sensor device

453:First Orientation Axis

454: Second infrared sensor device

455:Second Orientation Axis

510: curve

520: curve

610: curve

620: high frequency part

630: low frequency part

640: curve

710: curve

720: curve

800: flow chart

801: Operation

810: operation

820: Operation

830: Operation

840: Operation

850: operation

860: operation

870: Operation

880: operation

900: flow chart

901: Operation

910: Operation

920: Operation

930: Operation

932:Operation

934:Operation

936: Operation

940: Operation

942:Operation

950: operation

960: Operation

962:Operation

970: Operation

980: operation

990: operation

1000: electrochromic device

1002: Substrate

1004: first conductive layer (CL)

1006: Electrochromic layer (EC)

1008: ion conducting layer (IC)

1010: opposite electrode layer (CE)

1014: second conductive layer (CL)

1016: Voltage source

1020:Electrochromic stack

1100: Electrochromic device

1102: Substrate

1104: conductive layer (CL)

1106: Tungsten oxide electrochromic layer (EC)

1108: ion conducting layer (IC)

1110: nickel tungsten oxide opposite electrode layer (CE)

1114: conductive layer (CL)

1116: Power

1120:Electrochromic stack

1200: Electrochromic device

1250: window controller

1255: Microprocessor

1260: Pulse Width Modulator

1265: Signal conditioning module

1270: Computer-readable media

1275: Configuration file

1280: network

1300: Building Management Systems (BMS)

1301:Buildings

1302: Main window controller

1303: Main network controller

1305a: intermediate network controller

1305b: intermediate network controller

1310: End or Leaf Controller

1400: system

1401: EC device

1402: Main window controller

1403: Primary Network Controller

1405: intermediate network

1410: Communication Network/Leaf or End Window Controller

1412: Multi-sensor device

1490: Optional wall switch

1500: room

1501: table

1502: Infrared Cloud Detector System

1505: Electrochromic window

1510: Light sensor

1520: Overhang

1530: Infrared Cloud Detector

1532: shell

1534: Infrared sensor

1536: Ambient temperature sensor

1550: local window controller

1600: Flowchart

1610: Operation

1620: Operation

1630: Operation

1640: Operation

1650: Operation

1770: Operation

1780: Operation

1790: Operation

1800: Flowchart

1810: Operation

1820: Operation

1830: Operation

1840: Operation

1850: Operation

1860: Operation

1870: Operation

1900: Flowchart

1910: Operation

1920: Operation

1930: Operation

1932: Operation

1934: Operation

1936: Operation

1940: Operation

1942: Operation

1950: Operation

1970: Operation

1980:Operation

1990:Operation

1995: Operation

2000: Room

2001: table

2005: Electrochromic windows

2010: Visible light sensor

2020: Overhangs

2030: Multi-sensor devices

2032: shell

2034: Infrared sensor device

2050: window controller

2090: infrared radiation

2100: Flowchart

2110: Operation

2120: Operation

2130: Operation

2140: Operation

2150: Operation

2220: block

2270: Operation

2280: Operation

2290: Operation

2295: Operation

2300: flow chart

2310: Operation

2320: Operation

2330: Operation

2340: Operation

2350: Operation

2360: Operation

2370: Operation

2380: Operation

2400: flow chart

2405: Operation

2412: Operation

2414: Operation

2422: Operation

2424: Operation

2430: Operation

2432: Operation

2434: Operation

2436: Operation

2440: Operation

2490: Operation

2492:Operation

2496:Operation

2600: flow chart

2610: Operation

2620: Operation

2630: Operation

2640: Operation

2650: Operation

2700: Flowchart

2710: Operation

2720: Operation

2730: Operation

2740: Operation

2750: Operation

2800: Flowchart

2810: Operation

2820: Operation

2830: Operation

2832: Operation

2840: Operation

2850: Operation

2930: first curve

2932: Second Curve

3000: flow chart

3020: Operation

3025: Operation

3030: Operation

3050: Operation

3070: Operation

3100: Flowchart

3110: Operation

3120: Operation

3130: Operation

3140: Operation

3150: Operation

3160: Operation

3170: Operation

3180: Operation

3200: Infrared Cloud Detector System

3201: Multi-sensor device

3210: shell

3211: cover

3212: The thinning part

3214: base part

3220: Mast

3222: The first ambient temperature sensor

3301: Multi-sensor device

3302: shell

3304: Diffuser

3340: Visible light sensor

3342: Visible Light Sensor/Axis

3343: photosensitivity zone

3360: Third infrared sensor device

3361: Imaginary axis

3372: The first infrared sensor device

3373: imaginary axis

3374:Second Infrared Sensor Device

3375: imaginary axis

Figure 1 shows a schematic representation of a side view of an infrared cloud detector system according to some implementations.

2A shows a graph with two curves of temperature readings taken by the infrared sensors of the infrared cloud detector over time according to this implementation.

2B shows a graph with two plots of ambient temperature readings taken by the ambient temperature sensor of the infrared cloud detector discussed with respect to FIG. 2A over time.

2C shows a graph with two curves of the calculated difference between the temperature readings taken by the infrared sensor and the ambient temperature readings taken by the ambient temperature sensor of the infrared cloud detector discussed with respect to FIGS. 2A and 2B .

3 depicts a schematic diagram (side view) of an infrared cloud detector system including an infrared cloud detector and a light sensor according to an implementation.

4A shows a perspective view of a diagrammatic representation of an infrared cloud detector system including infrared cloud detectors in the form of multiple sensors, according to an implementation.

4B shows another perspective view of an infrared cloud detector system including an infrared cloud detector in the form of the multi-sensor shown in FIG. 4A .

Figure 4C shows a perspective view of some of the internal components of the multi-sensor device of the infrared cloud detector system shown in Figures 4A and 4B .

5A is a graph with a plot of intensity readings taken by a visible light photosensor over time.

5B is a graph with a plot of the difference between temperature readings taken by the infrared sensor and temperature readings taken by the ambient temperature sensor over time.

6A is a graph with a plot of intensity readings taken by a visible light photosensor over time.

6B is a graph with a plot of the difference between temperature readings taken by an infrared sensor over time and temperature readings taken by an ambient temperature sensor over time.

7A is a graph with a plot of intensity readings taken by a visible light photosensor over time.

7B is a graph with a plot of the difference between temperature readings taken by the infrared sensor and temperature readings taken by the ambient temperature sensor over time.

8 shows a flowchart describing a method of determining cloud cover conditions using temperature readings from an infrared sensor and an ambient temperature sensor, according to an implementation.

9 shows a flowchart describing a method of determining cloud cover conditions using readings from an infrared sensor, an ambient temperature sensor, and a light sensor of an infrared cloud detector system, according to an implementation.

Figure 10 depicts a schematic cross-section of an electrochromic device.

11A depicts a schematic cross-section of an electrochromic device in (or transitioning to) a bleached state.

FIG. 11B depicts a schematic cross-section of the electrochromic device shown in FIG. 11A but in (or transitioning to) a colored state.

Figure 12 depicts a simplified block diagram of components of a window controller according to one embodiment.

Figure 13 depicts a schematic diagram of an embodiment of a BMS according to an embodiment.

14 is a block diagram of components of a system for controlling the function of one or more tintable windows of a building, according to an embodiment.

15A shows the penetration depth of direct sunlight into a room through an electrochromic window between the exterior and interior of a building containing the room, according to an implementation.

15B shows direct sunlight and radiation entering a room through an electrochromic window under clear sky conditions, according to an implementation.

15C shows radiant light from the sky as may be blocked by or reflected from objects such as clouds and other buildings, according to an implementation.

16 depicts a flowchart showing the general control logic of a method for controlling one or more electrochromic windows in a building according to an embodiment.

Figure 17 is a diagram showing a particular implementation of one of the blocks from Figure 16 , according to an implementation.

Figure 18 depicts a flow diagram showing one particular implementation of the control logic for the operation shown in Figure 16 , according to an embodiment.

Figure 19 is a flow diagram depicting one particular implementation of the control logic for the operation shown in Figure 18 , according to an implementation.

20A shows the penetration depth of direct sunlight into a room through an electrochromic window between the exterior and interior of a building containing the room, according to an implementation.

20B shows direct sunlight and radiation entering a room through an electrochromic window under clear sky conditions, according to an implementation.

20C shows radiant light from the sky as may be blocked by or reflected from objects such as clouds and other buildings, according to an implementation.

Figure 20D shows infrared radiation from the sky, according to an implementation.

21 includes a flowchart depicting the general control logic of a method for controlling one or more electrochromic windows in a building according to an embodiment.

FIG. 22 includes a flow diagram of logic according to one implementation of one block of the flow diagram illustrated in FIG. 21 .

23 includes a flowchart depicting the control logic of module D' for determining a filtered infrared sensor value, according to an implementation.

24 includes a flowchart depicting control logic for making coloring decisions based on infrared sensor and/or light sensor data depending on whether it is during the morning range, during the day time range, during the evening range, or during the night, according to an implementation.

25 is an example of an occupancy lookup table according to certain aspects.

26 includes a flowchart depicting control logic for determining hue levels from module D when the current time is during the daytime range, according to certain aspects.

27 includes a flowchart depicting control logic for determining hue levels from module D when the current time is during the evening range, according to certain aspects.

28 includes a flowchart depicting control logic for determining hue levels from module C1 and/or module D when the current time is during the daytime range, according to certain aspects.

29 shows a graph of filtered infrared sensor values in millidegrees versus time during a 24-hour period, according to an implementation.

30 includes a flowchart depicting the control logic of module C1 for determining a tint level for one or more electrochromic windows in a building, according to an implementation.

FIG. 31 includes a flowchart depicting the control logic of module C1 ′ for determining a filtered light sensor value, according to an implementation.

32A shows a perspective view of a diagrammatic representation of an infrared cloud detector system with a multi-sensor device according to an implementation.

Figure 32B shows another perspective view of the multi-sensor device shown in Figure 32A .

Figure 33A shows a perspective view of components of the multi-sensor device shown in Figure 32A , according to an implementation.

Figure 33B shows another perspective view of the components of the multi-sensor device shown in Figure 32A .

I. Introduction

At certain times of the day, the intensity of visible light is at low levels, such as in the morning around sunrise and in the evening just before sunset. A light sensor calibrated to measure the intensity of visible light (referred to herein as a "visible light sensor" or "light sensor" generally) does not detect direct sunlight, and its intensity measurements at these times of day may not be useful in determining cloud conditions. In some aspects, the cloud condition is judged to be one of: 1) a "clear" condition when the sky is clear with no or almost no clouds; 2) a "partly cloudy" condition; and 3) a "cloudy" or "overcast" condition when the sky is cloudy. That is, a visible light photosensor directed toward the sky at these times will measure low intensity values during "clear" conditions, "partly cloudy" conditions, and "overcast" conditions. Therefore, taken by the visible light sensor alone The resulting intensity measurements may not accurately distinguish between different cloud conditions at such times. If the intensity measurements from the visible light sensor are used alone to determine "cloudy" conditions in the evening just before sunset (eg, when the measured intensity level drops below a certain minimum value), false "cloudy" conditions can be detected. Similarly, visible light sensor measurements are not valid for distinguishing between "cloudy" and "clear" conditions immediately before sunrise when there is no direct sunlight. During either of these time periods, light sensor measurements can be used to detect false "cloudy" conditions. Controllers that rely on false "cloudy" determinations from such light sensor readings can therefore implement inappropriate control decisions based on such false "cloudy" determinations. For example, a window controller controlling tint levels in an east-facing optically switchable window (e.g., an electrochromic window) may inappropriately clear the window, allowing direct glare from the rising sun to illuminate the room if the light sensor readings determine false "cloudy" conditions at a time immediately before sunrise.

Furthermore, controllers making decisions based primarily on current readings from visible light sensors do not take into account historical intensity levels in geographic areas that may be related to likely current/future cloud coverage conditions, eg, making control commands in anticipation of likely conditions. For example, on mornings when few clouds pass over the geographic area, there may be historically low light levels. In this case, a small amount of cloud temporarily blocking sunlight to the light sensors will result in the same "cloudy" condition determination as when a large storm rolls into the area. In this case, the passage of a small amount of cloud could cause the controller to switch the tintable window, and possibly lock the optically switchable window to an inappropriately low tint level, until the window can be switched to a higher (darker) tint level.

II. Infrared (IR) Cloud Detector System

Both clouds and water vapor absorb and re-emit radiation in discrete frequency bands on the infrared (IR) spectrum. Because clouds absorb and re-emit IR radiation and clear skies transmit IR radiation, clouds are generally warmer (have a higher temperature) than clear skies. In other words, the presence of clouds generally produces an enhanced IR signal (which corresponds to a roughly black body spectrum at about ground temperature) higher than that from a clear sky. There is also a small effect of atmospheric humidity, which can also produce an enhanced IR signal, particularly at low elevations. Based on these distinctions, devices that measure IR radiation can effectively detect cloud conditions.

Various implementations relate to infrared cloud detectors and methods thereof for detecting cloud cover or other cloud conditions based on infrared readings. Infrared cloud detectors typically include at least one infrared (IR) sensor and at least one ambient temperature sensor that are used in combination to obtain a temperature reading of the sky that can be used to detect cloud cover conditions. In general, the amount of infrared radiation emitted by a medium/object and then measured by an IR sensor varies depending on the temperature of the medium/object, the surface and other physical characteristics of the medium/object, the field of view of the IR sensor, and the distance between the medium/object and the IR sensor. The IR sensor converts IR radiation received within its field of view to a voltage/current, and converts the voltage/current to a corresponding temperature reading (eg, a digital temperature reading) of the medium/object within its field of view. For example, an IR sensor directed (orientated) to face the sky outputs temperature readings for the area of sky within its field of view. An IR sensor may be oriented in a particular direction (eg, azimuth and elevation) to preferentially capture IR radiation in a geographic region of the sky within its field of view centered around that direction. The ambient temperature sensor measures the temperature of the ambient air surrounding the sensor. Typically, an ambient temperature sensor is positioned to measure the temperature of the ambient air surrounding the infrared cloud detector. The infrared cloud detector also has a processor that determines the difference between the temperature readings taken by the IR sensor and the ambient temperature sensor, and uses this difference to detect the amount of cloud cover in the area of the sky within the field of view of the IR sensor.

Typically, temperature readings taken by ambient temperature sensors tend to fluctuate over a smaller range with changing weather conditions than sky temperature readings taken by infrared radiation sensors. For example, sky temperature readings taken by infrared radiation sensors tend to fluctuate at high frequencies during "intermittent cloudy" conditions in fast-moving weather patterns. Some implementations of the infrared cloud detector have logic to determine the difference (delta (Δ)) between the infrared sensor sky temperature reading ( T _sky ) and the ambient temperature reading ( T _amb ) according to Equation 1 to help normalize any fluctuations in the infrared sensor temperature reading ( T _sky ). In one example, if the difference (△) is determined to be above an upper threshold (e.g., about 0 millidegrees Celsius), then a "cloudy" condition is logically asserted, if the difference (△) is determined to be below a lower threshold (eg, about -5 millidegrees Celsius), then a "clear" condition is logically asserted, and if the difference (Δ) is determined to be between the upper and lower thresholds, an "intermittent cloudy" condition is logically asserted. In another example, a "cloudy" condition is logically asserted if the difference (Δ) is above a single threshold, and a "clear" condition is logically asserted if the delta (Δ) is below the threshold. In one aspect, the logic may apply one or more correction factors to the delta (Δ) before determining whether it is above or below a threshold. Some examples of correction factors that may be used in an implementation include humidity, sun angle/elevation, and site elevation. For example, a correction factor may be applied based on the altitude and density of the clouds being detected. Lower altitudes and/or higher density clouds correlate more closely with ambient temperature readings than with infrared sensor readings. Higher altitudes and/or less dense clouds correlate more closely with infrared sensor readings than with ambient temperature readings. In this example, a correction factor that gives higher weight to ambient temperature readings for lower altitudes and/or higher density clouds may be applied, or a correction factor that gives higher weight to infrared sensor readings for higher altitudes and/or less dense clouds may be used. In another example, a correction factor may be applied based on humidity and/or sun position to more accurately describe cloud cover and/or remove any outliers. The following describes the technical advantages of using the delta (Δ) to determine cloud conditions with reference to FIGS . 2A to 2C .

Since temperature readings are generally independent of the presence of direct sunlight, in some cases cloud cover conditions can be detected more accurately using temperature readings than visible light sensors can detect when the intensity of sunlight is low (e.g., in the early morning immediately before and immediately after sunrise, and in the early evening before sunset). During these times, the visible light sensor can potentially detect false "cloudy" conditions. According to such implementations, infrared cloud detectors can be used to detect cloud cover with accuracy independent of whether the sun is out or otherwise low light intensity levels exist, such as immediately before sunrise or sunset. In such implementations, relatively low temperatures generally indicate the likelihood of "clear" conditions, and relatively high temperature readings generally indicate the likelihood of "cloudy" conditions (ie, cloud cover).

In various implementations, the IR sensor of the infrared cloud detector is calibrated to measure the radiant flux of long wavelength infrared radiation within a specific range. The IR sensor's processor or a separate processor can be used to infer the temperature reading from these measurements. In one aspect, the IR sensor is calibrated to detect Infrared radiation is measured in the wavelength range between about 8 μm and about 14 μm. In another aspect, the IR sensor is calibrated to detect infrared radiation having a wavelength in excess of about 5 μm. In another aspect, the IR sensor is calibrated to detect infrared radiation in the wavelength range between about 9.5 μm and about 11.5 μm. In another aspect, the IR sensor is calibrated to detect infrared radiation in a wavelength range between about 10.5 μm and 12.5 μm. In another aspect, the IR sensor is calibrated to detect infrared radiation in the wavelength range between about 6.6 μm and 20 μm. Some examples of types of IR sensors that may be used include infrared thermometers (eg, thermopiles), infrared radiometers, infrared ground radiometers, infrared pyrometers, and the like. One commercially available example of an IR sensor is the Melexis MLX90614 manufactured by Melexis of Detroit, Michigan. Another commercially available example of an IR sensor is the TS305-11C55 temperature sensor manufactured by TE connectivity Ltd., Switzerland. Another commercially available example of an IR sensor is the SI-111 Infrared Radiometer manufactured by Apogee Temperature Sensors manufactured by TE connectivity Ltd., Switzerland.

In various implementations, the infrared cloud detector has an IR sensor positioned and oriented such that its field of view receives infrared radiation from a particular region of the sky of interest. In one implementation, the IR sensor may be located on the roof of a building and oriented with its sensing surface facing vertically upwards, or at a slight angle from vertical, such that its field of view has an area of sky above or at a distance from the building.

In some implementations, the infrared cloud detector has a protective housing, and the infrared sensor is located within the housing. The housing may have a cover with one or more apertures or thinned areas that allow/restrict transmission of infrared radiation to the infrared sensor. In some cases, the cover may be formed of plastic, such as polycarbonate, polyethylene, polypropylene, and/or thermoplastics such as nylon or other polyamides, polyesters, or other thermoplastics, plus other suitable materials. In one example, the material is weather resistant plastic. In other cases, the cover may be formed from a metallic material such as aluminum, cobalt, or titanium, or a semi-metallic material such as nylon mixed with aluminum powder (alumide). In some implementations, the cover can be sloped or convex to prevent accumulation of water. Depending on the type of material used to form the cover, the cover may 3D printed, injection molded, or formed via another suitable process or processes.

In some implementations, the cover includes one or more apertures or thinned regions to increase transmission (reduce blockage) of incident radiation or other signals to detectors within the housing. For example, the cover may include one or more apertures or thinned regions proximate the infrared sensor in the housing to allow improved transmission of incident infrared radiation to the infrared sensor. Apertures or thinned regions may also improve the transmission of other signals (eg, GPS signals) to other detection devices within the housing. Additionally or alternatively, some or all of the covers may be formed from a light diffusing material. In some implementations, the cover can be connected to the housing via adhesive or by some mechanical coupling mechanism, such as by using threads and threads through or via a pressure seal or other compression fit.

The field of view of the sensing surface of an infrared sensor is defined by its material composition and its structure. In some cases, the field of view of the infrared sensor may be narrowed by obstacles. Some examples of obstacles include building structures such as overhangs or roof structures, obstacles in the vicinity of buildings such as trees or another structure, and the like. As another example, if the infrared sensor is located within the housing, structures within the housing can narrow the field of view.

In one aspect, a single IR sensor has a vertical unconstrained field of view from about 50 degrees to about 130 degrees +- 40 degrees from vertical. In one aspect, the IR sensor has a field of view in the range of 50 degrees and 100 degrees. In another aspect, the IR sensor has a field of view in the range of 50 degrees and 80 degrees. In another aspect, the IR sensor has a field of view of about 88 degrees. In another aspect, the IR sensor has a field of view of about 70 degrees. In another aspect, the IR sensor has a field of view of about 44 degrees. The field of view of an IR sensor is usually defined as a conical volume. IR sensors typically have a wider field of view than visible light sensors and are therefore able to receive radiation from a larger area of the sky. Since an IR sensor can take readings of a larger area of the sky, an IR sensor may be more useful in determining approaching conditions (eg, approaching storm clouds) than a visible light light sensor, which would be more limited to detecting current conditions affecting the immediate vicinity of the light sensor within its smaller field of view. In one aspect, a five-sensor blocking IR sensor configuration of mounted sensors (e.g., in a multi-sensor configuration) has four angled mounted IR sensors, each One constrained to a 20-70 degree or 110-160 degree field of view, and one upward facing IR sensor constrained to a 70-110 degree field of view.

Certain IR sensors tend to be more effective at measuring sky temperature when direct sunlight is not hitting the sensing surface. In some implementations, the infrared cloud detector has a structure that blocks direct sunlight from the sensing surface of the IR sensor, or has a structure that diffuses direct sunlight before it hits the sensing surface of the IR sensor (eg, a housing of opaque plastic). In one implementation, the IR sensor may be obscured by a building or overhanging the infrared cloud detector. In another implementation, the IR sensor may be located within a protective housing with diffuser material between the sensing surface of the IR sensor and the sky to diffuse any direct sunlight from reaching the sensing surface of the IR sensor and also provide protection from potentially harmful elements such as dust, animals, etc. Additionally or alternatively, some implementations only use IR sensor readings taken before sunrise or after sunset to avoid the possibility of direct sunlight hitting the IR sensor. In such implementations, light sensor readings or other sensor readings may be used to detect cloud cover conditions between sunrise and sunset.

In various implementations, the infrared cloud detector has an ambient temperature sensor for measuring the temperature of the air surrounding the ambient temperature sensor. Typically, an ambient temperature sensor is positioned in contact with the outdoor environment (eg, located outside a building) to take a temperature reading of the sky. The ambient temperature sensor may be, for example, a thermistor, thermocouple, resistance thermometer, thermocouple, silicon bandgap temperature sensor, or the like. One commercial example of an ambient temperature sensor that can be used is the Pt100 thermometer probe manufactured by Omega. Some implementations include an ambient temperature sensor positioned to avoid direct sunlight hitting its sensing surface. For example, the ambient temperature sensor may be located under an overhang or mounted under a structure that shields the ambient temperature sensor from direct sunlight.

While many implementations of infrared cloud detectors described herein include one IR sensor and one ambient temperature sensor, it will be understood that other implementations may include more than one IR sensor and/or more than one ambient temperature sensor. For example, in one implementation, the infrared cloud detector includes two or more IR sensors for redundancy and/or to direct the IR sensors to different regions of the sky. Other Additionally or alternatively, in another implementation, the infrared cloud detector may have two or more ambient temperature sensors for redundant use. An example of a system for detecting clouds using different regions of the sky directed by two IR sensors can be found in International Application PCT/US15/53041, filed September 29, 2015 and entitled "SUNLIGHT INTENSITY OR CLOUD DETECTION WITH VARIABLE DISTANCE SENSING," which is hereby incorporated by reference in its entirety.

Various implementations of infrared cloud detectors have the basic functionality of detecting cloud cover conditions. In some cases, infrared cloud detectors can detect "cloudy" conditions as well as "clear" conditions. Additionally, some implementations may further break down "cloudy" conditions into classes. For example, an implementation may classify "cloudy" conditions into "overcast" or "intermittent clouds." In another example, an implementation may assign different levels (eg, 1-10) of cloudiness to the "cloudy" condition. In yet another example, an implementation can predict future cloud conditions, ie, the likelihood of cloud conditions occurring at a future time. Additionally or alternatively, some implementations may detect other weather conditions as well.

In various implementations, the infrared cloud detector includes an IR sensor configured to take a sky temperature reading T _sky , and an ambient temperature sensor configured to take an ambient temperature reading T _amb . The infrared cloud detector also includes one or more processors containing program instructions executable to perform various functions of the infrared cloud detector. The processor executes programmed instructions to determine the temperature difference between the temperature readings, Delta (Δ), as provided in Equation 1 . The processor also executes programmed instructions to determine cloud cover conditions based on the delta (Δ). As mentioned above, in some cases, using the ambient temperature reading can help normalize any rapid fluctuations in the IR sensor temperature reading.

Difference (△) = Infrared sensor sky temperature reading (T _sky ) - ambient temperature reading (T _amb ) (Equation 1)

In one implementation, the processor executes programmed instructions to compare the delta (Δ) with an upper threshold and a lower threshold, and determine a cloud coverage condition. If the difference (△) is higher than the upper threshold, then judge "Sunny" conditions. If the difference (△) is lower than the lower threshold, the "cloudy" condition is determined. If the difference (Δ) is below the upper threshold and above the lower threshold (ie, between the thresholds), then an "intermittent" cloud cover condition is determined. Additionally or alternatively, additional factors may be used to determine cloud cover conditions when the difference (Δ) is between thresholds. This implementation works well in the morning around dawn and in the evening around dusk to accurately determine "cloudy" or "clear" conditions. Between sunrise and sunset, additional factors may be used to determine cloud cover conditions, such as by using visible light sensor values. Some examples of additional factors include: elevation, wind speed/direction, and sun elevation/angle.

A. Infrared (IR) sensor cloud detection system

Figure 1 shows a schematic representation of a side view of a system with an infrared cloud detector 100 , according to some implementations. The infrared cloud detector 100 has a housing 101 with a cover 102 having an aperture or thinned portion 104 on a first surface 106 of the housing 101 . The housing 101 also has a second surface 108 opposite to the first surface 106 . Infrared cloud detector 100 also includes: an IR sensor 110 configured to take a temperature reading T _sky based on infrared radiation received within its cone of field of view 114 ; an ambient temperature sensor 130 for taking an ambient temperature reading T _amb ; and a processor 140 in communication (wired or wireless) with IR sensor 110 and ambient temperature sensor 130 . In one aspect, the IR sensor is one of an infrared thermometer (eg, a thermopile), an infrared radiometer, an infrared ground radiometer, and an infrared pyrometer. In one aspect, the ambient temperature sensor is one of a thermistor, thermometer, and thermocouple.

In FIG. 1 , the IR sensor 110 is located behind the aperture or thinned portion 104 and within the housing of the housing 101 . The aperture or thinned portion 104 enables the IR sensor 110 to measure infrared radiation transmitted through the aperture or thinned portion 104 and received at its sensing surface. The IR sensor 110 includes an imaginary axis 112 that is normal to the sensing surface of the IR sensor 110 and passes through the center of the IR sensor 110 . In the illustrated example, the IR sensor 110 is oriented such that its axis 112 is in a vertical orientation and the sensing surface faces upward. In other examples, the IR sensor 110 may be directed such that the sensing surface faces another orientation to, for example, direct the IR sensor to a particular region of the sky. The IR sensor 110 has a cone-shaped field of view 114 out of the housing 102 through the aperture or thinned portion 104 . In this example, the portion of the cover 102 surrounding the aperture or thinned portion 104 is made of a material that blocks infrared radiation, and the perimeter of the aperture or thinned portion 104 defines a field of view 114 . The field of view 114 has an angle α and is centered about the axis 112 . In FIG. 1 , the ambient temperature sensor 130 is positioned and attached to the second surface 108 of the housing 102 away from the edge to avoid direct sunlight hitting the ambient temperature sensor 130 when the infrared cloud detector 100 is in this orientation. Although not shown, the infrared cloud detector 100 also includes one or more structures that hold the infrared sensor 110 and other components in place within the housing 101 .

The infrared cloud detector 100 also has the logic to calculate the difference (Δ) between the infrared sensor sky temperature reading ( T _sky ) and the ambient temperature reading ( T _amb ) at each reading time and determine the cloud coverage condition based on the calculated difference (Δ). During operation, IR sensor 110 takes a sky temperature reading T _sky based on infrared radiation received from a region of the sky within its field of view 114 , and ambient temperature sensor 130 takes an ambient temperature reading T _amb of the ambient air surrounding infrared cloud detector 100 . The processor 140 receives a signal from the IR sensor 110 with a temperature reading T _sky and a signal from the ambient temperature sensor 130 with an ambient temperature reading T _amb . Processor 140 executes instructions stored in memory (not shown) that use logic that calculates a difference (Δ) between the infrared sensor temperature reading ( T _sky ) and the ambient temperature reading ( T _amb ) at a particular time to determine cloud cover conditions. Processor 140 may execute instructions to: if the difference (Δ) at that time is above the upper threshold, then determine a "cloudy" condition, if the difference (Δ) is below the lower threshold, then determine a "clear" condition, and if the difference (Δ) is determined to be between the upper and lower thresholds, then determine an "intermittently cloudy" condition. Processor 140 may also execute instructions stored in memory to perform other operations of the methods described herein.

Although a single infrared sensor 110 is illustrated in FIG. 1 , in another implementation, two or more infrared sensors may be used redundantly should one fail and/or be obscured by, for example, bird droppings or another environmental agent. In one implementation, two or more infrared sensors are used facing different orientations to capture IR radiation from different fields of view and/or at different distances from the building/structure. If two or more IR sensors are located within the housing of infrared cloud detector 100 , the IR sensors are typically offset from each other by a distance sufficient to reduce the likelihood that an obscuring agent will affect all IR sensors. For example, the IR sensors can be separated by at least about one inch or at least about two inches.

B. Comparison of infrared sensor temperature readings, ambient temperature readings and delta values during a sunny day and a cloudy day in the afternoon

As discussed above, sky temperature readings taken by ambient temperature sensors tend to fluctuate over a smaller range than sky temperature readings taken by infrared radiation sensors. Some implementations of the infrared cloud detector have logic to determine the difference (delta (Δ)) between the infrared sensor temperature reading ( T _sky ) and the ambient temperature reading ( T _amb ) according to Equation 1 to help normalize any fluctuations in the infrared sensor temperature reading ( T _sky ). By way of comparison, FIGS. 2A - 2C contain graphs of temperature readings T _IR taken by the infrared sensor of the infrared cloud detector according to one implementation, sky temperature readings T _sky taken by the ambient temperature sensor of the infrared cloud detector, and examples of the difference (Δ) between these readings. Each graph contains two curves: a curve for readings taken during a sunny day and a curve for readings taken during a cloudy day in the afternoon. The infrared cloud detector used in this example included components similar to those described with respect to the infrared cloud detector 100 shown in FIG. 1 . In this case, the infrared cloud detector is located on the roof of the building and the infrared sensor is oriented to face vertically upwards. The infrared sensor is calibrated to measure infrared radiation in the wavelength range from about 8 μm to about 14 μm. To protect the infrared sensor from direct sunlight, the infrared sensor is located behind a cover formed of a light diffusing material such as plastic (eg, polycarbonate, polyethylene, polypropylene, and/or thermoplastics such as nylon or other polyamides, polyesters, or other thermoplastics, plus other suitable materials). In this example, the infrared cloud detector also has logic to calculate the difference (difference (Δ) ) between the sky temperature reading T _sky taken by the IR sensor and the ambient temperature reading Tamb taken by the ambient temperature sensor of the infrared _cloud detector. The logic can also be used to determine a "cloudy" condition if the difference (Δ) is at or above the upper threshold, a "clear" condition if the difference (Δ) is at or below the lower threshold, and an "intermittently cloudy" condition if the difference (Δ) is determined to be between the upper and lower thresholds.

FIG. 2A shows a graph with two curves of temperature readings T _sky taken by the infrared sensor of the infrared cloud detector over time according to this implementation. Each of the two curves has temperature readings T _sky taken by the infrared sensor over a time period of one day. The first curve 110 has temperature readings T _sky taken by the infrared sensor during the first day with cloudy afternoons. The second curve 112 has the temperature readings T _sky taken by the infrared sensor during the second day when the sky is clear. As shown, the temperature reading T _sky of the first curve 110 taken during the afternoon of the first day with cloudy afternoons is generally higher than the temperature reading T _sky of the second curve 112 taken during the second day of the day with all clear weather.

2B shows a graph with two plots of ambient temperature readings T _amb over time taken by the ambient temperature sensor of the infrared cloud detector discussed with respect to FIG. 2A . Each of the two curves has temperature readings T _amb taken by the ambient temperature sensor over a time period of one day. To avoid direct sunlight hitting the ambient temperature sensor, shield it from direct sunlight. The first curve 220 has the temperature readings taken by the ambient temperature sensor during the first day with cloudy afternoons. The second curve 222 has the temperature readings taken by the infrared sensor during the second day when the sky was clear. As shown, the ambient temperature reading Tamb of the first curve 220 _taken during the first day with cloudy afternoons is at a lower level than the temperature reading _Tamb of the second curve 222 taken during the second day of the clear day.

2C shows a graph with two curves of the calculated difference (Δ) between _the sky temperature reading T _sky taken by the IR sensor and the ambient temperature reading Tamb taken by the ambient temperature sensor of the infrared cloud detector discussed with respect to FIGS . 2A and 2B . Each of the two curves has a calculated difference (Δ) over a time period of the day. The first curve 230 is the calculated delta (Δ) of readings taken during the first day with cloudy afternoons. The second curve 232 is the calculated delta ([Delta]) taken during the second day of full day clear. The graph also includes an upper threshold and a lower threshold.

In FIG. 2C , the value of the difference (Δ) of the second curve 232 during the time interval from immediately before sunrise to immediately after sunrise and during the time interval from immediately before sunset to sunset is below the lower threshold. Using the calculated delta (Δ) values shown in the graph in Figure 2C , the logic of the infrared cloud detector will determine a "clear" condition during this time interval. Also, since the value of the delta (Δ) of the second curve 232 is below the lower threshold value at most other times of the day, the logic of the infrared cloud detector will also determine a "clear" condition for those other times.

In FIG. 2C , the value of the delta (Δ) of the first curve 230 is above the upper threshold for most of the afternoon, and the infrared cloud detector will determine "cloudy" conditions during the afternoon. The value of the delta (Δ) of the first curve 230 is below the lower threshold value during a time interval from immediately before sunrise to immediately after sunrise and during a time interval from immediately before sunset to sunset. Based on these calculated delta (Δ) values, the logic of the infrared cloud detector will determine "clear" conditions during this time interval. During the brief period of time in the transition between early afternoon and late afternoon, the value of the delta (Δ) of the first curve 230 is between the lower threshold and the upper threshold. Based on these calculated delta (Δ) values, the logic of the infrared cloud detector will determine an "intermittently cloudy" condition.

C. Infrared cloud detector system with light sensor

In some implementations, the infrared cloud detector system also includes a visible light sensor (eg, a photodiode) for measuring the intensity of visible radiation during operation. These systems typically include at least one infrared sensor, at least one ambient temperature sensor, at least one visible light sensor, and logic for determining a cloud cover condition based on readings taken by one or more of the infrared sensor, ambient temperature sensor, and visible light sensor. In some cases, the infrared sensor is calibrated to measure wavelengths in the 8 μm to 14 μm spectrum. In some cases, the light sensor is calibrated to detect the intensity of visible light in the bright light range (eg, between about 390 nm and about 700 nm). The light sensor can be located in/on the same housing as the IR sensor and the ambient temperature sensor, or it can be located separately. In some cases, the logic determines cloud cover conditions based on a calculated delta (Δ) value between the infrared sensor temperature reading T _sky and the ambient temperature reading Tamb , for example, when the _confidence level of the infrared sensor is high and/or the confidence level of the light sensor is low. The logic determines cloud cover conditions based on light sensor readings when the confidence level of the infrared sensor is low and/or the confidence level of the light sensor is high.

In various implementations, the infrared cloud detector system includes logic for determining cloud cover conditions using one or more of the following as inputs: time of day, day of year, temperature reading T _sky from the infrared sensor, ambient temperature reading _Tamb from the ambient temperature sensor, and light intensity reading from the light sensor, an oscillation frequency of the visible light intensity reading from the light sensor, and an oscillation frequency of the temperature reading T _sky from the infrared sensor. In some cases, the logic determines the frequency of oscillation from a visible light intensity reading, and/or determines the frequency of oscillation from a temperature reading T _sky . The logic determines whether the time of day is during one of four time periods: (i) a time period shortly before sunrise and to slightly after sunrise; (ii) a time period defined as after (i) and before (iii); (iii) a time period shortly before sunset (dusk) and until sunset; or (iv) a time period defined as after (iii) and before (i). In one instance, the time of sunrise can be determined from measurements taken by the visible wavelength light sensor. For example, time period (i) may end at the point when the visible light sensor starts measuring direct sunlight, ie, the point at which the intensity reading of the visible light sensor is at or above a minimum intensity value. Additionally or alternatively, time period (iii) may be determined to end at the point at which the intensity reading from the visible wavelength light sensor is at or below a minimum intensity value. In another example, the time of sunrise and/or the time of sunset can be calculated based on the day of the year using a solar calculator, and the time periods (i) and (iii) can be calculated by a defined period of time (e.g., 45 minutes) around the calculated time of sunrise/sunset. If the time of day is within the (i) or (iii) time period, the confidence level for light sensor readings tends to be low and infrared sensor readings tends to be high. In this case, the logic determines the cloud cover condition based on the delta (Δ) calculated with or without the correction factor. For example, the logic may determine a "cloudy" condition if the difference (Δ) is above the upper threshold, a "clear" condition if the difference (Δ) is below the lower threshold, and an "intermittently cloudy" condition if the difference (Δ) is between the upper and lower thresholds. As another example, the logic may assert a "cloudy" condition if the delta (Δ) is above a single threshold, and a "clear" condition if the delta (Δ) is below the threshold. If the time of day is during (ii) daylight, the confidence level for light sensor readings is at a high level, and the confidence level for infrared sensor readings tends to be low. In this case, the logic may use the light sensor readings to determine cloud cover conditions as long as the calculated difference between the infrared readings and the light sensor readings is at or below an acceptable value. For example, the logic may determine a "clear" condition if the light sensor reading is above a certain intensity level, and a "cloudy" condition if the light sensor reading is at or below the intensity level. If the calculated difference between the infrared reading and the light sensor reading increases above an acceptable value, the confidence level of the infrared reading is increased and the logic determines cloud cover conditions based on the difference (Δ) as described above. Alternatively or additionally, if the light sensor reading is determined to oscillate at a frequency greater than the first defined level, the confidence level for the infrared reading is increased and the logic determines cloud cover conditions based on the delta (Δ). If the infrared reading is determined to oscillate at a frequency greater than the second defined level, the confidence level for the light sensor reading is increased, and the logic determines a cloud cover condition based on the light sensor reading. If the time of day is during (iv) nighttime, the logic may determine cloud cover conditions based on the delta (Δ) as described above. Other implementations of logic that may be used by an infrared cloud detector system are described herein , including various logic described with reference to FIGS .

3 depicts a schematic diagram (side view) of an infrared cloud detector system 300 including an infrared cloud detector 310 and an external visible light sensor 320 according to one implementation. The infrared cloud detector 310 includes a housing 312 , an infrared sensor 314 inside the casing of the casing 312 and an ambient temperature sensor 316 also inside the casing of the casing 312 . Infrared sensor 314 is configured to take a temperature reading T _sky based on infrared radiation received from a region of the sky within its cone-shaped field of view 315 . The ambient temperature sensor 316 is configured to take an ambient temperature reading T _amb of the ambient air surrounding the infrared cloud detector 310 . In one aspect, the infrared sensor 314 is one of an infrared thermometer (eg, a thermopile), an infrared radiometer, an infrared ground radiometer, and an infrared pyrometer. In one aspect, the ambient temperature sensor is one of a thermistor, thermometer, and thermocouple.

Infrared cloud detector 310 is shown on the roof of a building having room 330 with a tintable window 332 (eg, an electrochromic window with at least one electrochromic device), and external visible light sensor 320 is on an exterior surface of the building. Tintable window 332 is located between the exterior and interior of the building containing room 330 . FIG. 5 also shows a table 334 in room 330 . Although in this example the light sensor 320 is located separately from the infrared cloud detector 310 , in other implementations the light sensor 320 is located in the housing of the housing or on the outside of the housing 312 .

The infrared sensor 314 includes an imaginary axis that is perpendicular to the sensing surface of the infrared sensor 314 and passes through its center. The infrared cloud detector 310 is supported by a wedge-shaped structure that orients the infrared cloud detector 310 so that its axis is directed at an inclination angle β from the horizontal. In other implementations, other components may be used to support the infrared cloud detector 310 . Infrared sensor 314 is directed such that the sensing surface faces the sky and can receive infrared radiation from the region of sky within its field of view 315 . The ambient temperature sensor 130 is located within the casing of the housing 312 , away from the edge, and is covered by an overhang of the housing 312 to prevent direct sunlight from irradiating the sensing surface of the ambient temperature sensor 130 . Although not shown, infrared cloud detector 310 also includes one or more structures that hold its components within housing 312 .

In FIG. 3 , infrared cloud detector system 300 also includes a controller 340 having a processor that executes instructions stored in memory (not shown) for using the logic of infrared cloud detector system 300 . Controller 340 communicates (wirelessly or wired) with infrared sensor 314 and ambient temperature sensor 316 to receive signals with temperature readings. Controller 340 is also in communication (wireless or wired) with light sensor 320 to receive a signal with a visible light intensity reading.

In some implementations, power/communication lines may run from a building or another structure to infrared cloud detector 310 . In one implementation, the infrared cloud detector 310 includes a network interface that couples the infrared cloud detector 310 to a suitable cable. Infrared cloud detector 310 may communicate data via a network interface to controller 340 or another controller of the building (eg, a network controller and/or a master controller). In some other implementations, infrared cloud detector 310 may additionally or alternatively include a wireless network interface to enable wireless communication with one or more external controllers.

In some implementations, infrared cloud detector 310 or other examples of infrared cloud detectors may also include a battery within or coupled to its housing to power the sensors and electrical components within it. Instead or in addition to power from a power supply (eg, from a building power supply), the battery can provide this power. In one implementation, an infrared cloud detector further includes at least one photovoltaic cell, for example, on an outer surface of the housing. The at least one photovoltaic cell may provide power instead or in addition to power provided by any other power supply.

Infrared cloud detector system 300 also includes logic for determining cloud cover conditions using one or more of the following as inputs: time of day, day of year, temperature reading T _sky from infrared sensor 314 , ambient temperature reading _Tamb from ambient temperature sensor 316 and light intensity reading from light sensor 320 , oscillation frequency of visible light intensity reading from light sensor 320 , and temperature reading from infrared sensor 314 Count the oscillation frequency of T _sky . During operation, infrared sensor 314 takes a temperature reading T _sky based on infrared radiation received from a region of the sky within its field of view 315 , ambient temperature sensor 316 takes an ambient temperature reading T _amb of the ambient air surrounding infrared cloud detector 310 , and light sensor 320 takes a reading of the intensity of visible light received at its sensing surface. The processor of the controller 340 receives the signal from the infrared sensor 314 having a temperature reading T _sky , the signal from the ambient temperature sensor 316 having an ambient temperature reading Tamb and the _signal from the light sensor 320 having an intensity reading. The processor executes instructions stored in memory for using logic to determine cloud cover conditions based on various inputs. An example of this logic is described above and also with reference to FIG. 9 . In one implementation, the controller 340 is also in communication with and configured to control one or more building components. For example, controller 340 may be in communication with tintable window 332 and configured to control its tint level. In this implementation, infrared cloud detector system 300 also includes logic for determining control decisions for one or more building components (eg, tintable windows 332 ) based on determined cloud coverage conditions. One example of logic for deciding control decisions based on determined cloud cover conditions is described in more detail with respect to FIG. 10 .

Although a single infrared sensor 314 , ambient temperature sensor 316 , and visible light sensor 320 are illustrated in FIG. 3 , it should be understood that in another implementation, the disclosure is not so limited and additional components may be used. For example, multiple components are available for redundant use should one component fail and/or be blocked or otherwise prevented from functioning. In another example, two or more components may be used in different locations or in different orientations to capture different information. In one implementation, two or more infrared sensors are used facing different orientations to capture infrared radiation from different fields of view and/or at different distances from the building/structure. Where there are multiple sensors, the cloud cover condition may be determined using an average or average of the values from the multiple sensors. If two or more IR sensors are located within the housing of the infrared cloud detector, the IR sensors are typically offset from each other by a distance sufficient to reduce the likelihood that an obscuring object will affect all IR sensors. For example, the IR sensors can be separated by at least about one inch or at least about two inches.

Further examples of infrared cloud detector systems in the form of multi-sensor devices are described in Section D below.

D. Multi-sensor device

According to various aspects, the infrared cloud detector system includes thermal sensors for measuring thermal radiation from the sky and ambient temperature of the environment. Outputs thermal sensor readings in degrees (eg, milli-degrees Celsius, degrees Celsius, degrees Fahrenheit, degrees Kelvin, etc.). Some examples of thermal sensor types that may be implemented include an infrared sensor for measuring the temperature of the sky ( T _sky ), an ambient temperature sensor for measuring the ambient temperature of the environment ( T _amb ), and an infrared sensor arrangement that includes both an onboard infrared sensor for measuring the temperature of the sky ( T _sky ) and an ambient temperature sensor for measuring the ambient temperature ( T _amb ). In implementations using an infrared sensor device having both an on-board infrared sensor and an on-board ambient temperature sensor, the device may output a reading of one or more of sky temperature ( T _sky ), ambient temperature ( T _amb ), and the difference between T _{sky and Tamb} ( Δ).

According to some aspects, the ambient temperature may be implemented as a thermocouple, thermistor, or the like. The ambient temperature sensor can be part of the infrared sensor or can be a separate sensor.

In certain implementations, an infrared cloud detector system includes an infrared cloud detector having one or more infrared sensors and one or more visible light sensors in the format of a multi-sensor device that may have various other optional sensors (e.g., ambient temperature sensors) and electrical components within or on its housing. Details of various examples of multi-sensor devices are described in U.S. Patent Application Nos. 15/287,646, filed October 6, 2016, and entitled "MULTI-SENSOR," and in U.S. Patent Application Nos. 14/998,019, filed October 6, 2015, and entitled "MULTI-SENSOR," which are hereby incorporated by reference in their entirety. The multi-sensor devices of these implementations are configured to be located in an environment outside the building so that the sensors are exposed to the outside environment, for example, on the roof of the building. In some of these implementations, power/communication lines run from the building to the multi-sensor device. In one such case, the multi-sensor device includes a network interface that couples the multi-sensor device to a suitable cable. The multi-sensor device can communicate sensor data and other information via a network interface to one or more external controllers, such as a local controller, a network controller, and/or a building master controller. In other implementations, the multi-sensor device may additionally or alternatively include a wireless network interface enabling wireless communication with one or more external controllers. In some implementations, the multi-sensor device may also include a battery within or coupled to its housing to power the sensors and electrical components within it. Instead or in addition to power from a power supply (eg, from a building power supply), the battery can provide this power. In some implementations, the multi-sensor device further includes at least one photovoltaic cell, eg, on a surface of its housing. As an alternative or in addition to power from another source, photovoltaic cells can provide power to the multi-sensor device.

- Instance A

4A , 4B and 4C show perspective views of a diagrammatic representation of an infrared cloud detector system 400 including an infrared cloud detector in the form of a multi-sensor device 401 according to one such implementation. 4A and 4B show that the multi-sensor device 401 includes a housing 410 coupled to a mast 420 . Mast 420 may serve as a mounting assembly including a first end portion for coupling to base portion 414 of housing 410 and a second end portion for mounting to a building. In one example, the base portion 414 is mechanically threaded therethrough or fixedly attached or otherwise coupled to or coupled with the first end portion of the mast 420 via a rubber gasket. Mast 420 may also include a second end portion, which may include a mounting or attachment mechanism for mounting or attaching mast 420 to the roof of a building (e.g., on the roof of a building having room 330 , as shown in FIG. 3 ), such as to a surface of the roof, a wall on the roof, or to another structure on the roof. The housing includes a cover 411 , which is depicted as being formed from a light diffusing material. Cover 411 also includes a thinned portion 412 . In other examples, cover 411 may be opaque or transparent.

FIG. 4B also shows that the multi-sensor device 401 includes an ambient temperature sensor 420 located on the bottom exterior surface of the base portion 414 . The ambient temperature sensor 420 is configured to measure the ambient temperature of the external environment during operation. The ambient temperature sensor 420 is located on the bottom surface to help shield it from direct solar radiation, for example, when the infrared cloud detector system 400 is located in an outdoor environment with the upper surface facing upward. The temperature sensor 420 may be, for example, a thermistor, thermocouple, resistance thermometer, silicon bandgap temperature sensor, or the like.

Figure 4C shows a perspective view of some of the internal components of the multi-sensor device 401 of the infrared cloud detector system 400 shown in Figures 4A and 4B . As shown, the infrared cloud detector system 400 further includes a visible light sensor 440 , redundant first infrared sensor means 452 and second infrared sensor means 454 . A first infrared sensor device 452 and a second infrared sensor device 454 are located on the upper portion of the multi-sensor device 401 and positioned behind a cover 411 (shown in FIGS. 4A and 4B ) formed of a light diffusing material.

As shown in FIG. 4C , the first infrared sensor device 452 has a first orientation axis 453 perpendicular to its sensing surface. The second infrared sensor device 454 has a second orientation axis 455 perpendicular to its sensing surface. In the illustrated example, the first infrared sensor device 452 and the second infrared sensor device 454 are positioned such that their orientation axes 453 , 455 face outward from the top portion of the housing 410 (shown in FIGS. 4A and 4B ) so that temperature readings during operation can be taken based on infrared radiation captured from above the multi-sensor device 401 . The first infrared sensor device 452 is separated from the second infrared sensor device 454 by at least about one inch. In one aspect, each infrared sensor device 452 , 454 has an infrared sensor for measuring the sky temperature ( T _sky ). In another aspect, each infrared sensor device 452 , 454 has an infrared sensor for detecting thermal radiation for measuring sky temperature ( T _sky ) and an on-board ambient temperature sensor for measuring ambient temperature ( T _amb ).

During operation, the first infrared sensor device 452 and the second infrared sensor device 454 detect infrared radiation radiated from any object or medium within their field of view to measure the sky temperature ( T _sky ). The field of view is based on the physical and material properties of the first infrared sensor 452 and the second infrared sensor 454 . Based solely on their physical and material properties, some examples of infrared sensors have a field of view ranging from about 50 degrees to about 80 degrees. In one particular example, the infrared sensor has a field of view of about 70 degrees.

The light sensor 440 has an orientation axis 442 perpendicular to its sensing surface. The light sensor 440 is positioned behind the thinned portion 412 of the housing 410 , as shown in FIG. 4A . The thinned portion 412 allows the light sensor 440 to receive visible light radiation through the thinned portion 412 . During operation, light sensor 440 measures the intensity of visible light received through thinned portion 412 .

Infrared cloud detector system 400 also includes logic for making decisions based on sensor data collected by multi-sensor device 401 . In this case, the multi-sensor device 401 and/or one or more external controllers (not shown) include memory and one or more processors that can execute instructions stored in the memory (not shown) for using the logic of the infrared cloud detector system 400 . The one or more external controllers communicate with multi-sensor device 401 (eg, via wireless or wired communication) to receive signals having sensor readings or filtered sensor values taken by infrared sensors 452 , 454 , ambient temperature sensor 420 , and light sensor 440 . In some implementations, power/communication lines may run from a building or another structure to infrared cloud detector system 400 . In one implementation, infrared cloud detector system 400 includes a network interface that can be coupled to a suitable cable. The infrared cloud detector system 400 can communicate data to one or more external controllers of the building via a network interface. In some other implementations, infrared cloud detector system 400 may additionally or alternatively include a wireless network interface to enable wireless communication with one or more external controllers. In some implementations, infrared cloud detector system 400 may also include a battery within or coupled to the housing to power the sensors and electrical components within it. Instead or in addition to power from a power supply (eg, from a building power supply), the battery can provide this power. In some implementations, the infrared cloud detector system 400 further includes at least one photovoltaic cell, eg, on a surface of the housing.

According to one aspect, infrared cloud detector system 400 includes logic for determining cloud cover conditions using one or more of the following as inputs: time of day, day of year, temperature reading T _sky from one or both of infrared sensor devices 452 , 454 , ambient temperature reading Tamb from ambient temperature sensor 420 and light intensity reading from light sensor 440 , oscillation of visible light intensity reading from light sensor 440 Frequency _and oscillation frequency of the temperature reading T _sky from the infrared sensor means 452 , 454 . Examples of this logic are described herein, for example, with respect to FIGS . 8-10 .

According to another aspect , infrared cloud detector system 400 includes various logic, such as some of the logic described with reference to FIGS . For example, in one implementation, multi-sensor device 401 and/or one or more external controllers include logic for: 1) determining a filtered infrared sensor value based on the temperature reading T _sky from one or both of infrared sensor devices 452 , 454 and the ambient temperature reading _Tamb from ambient temperature sensor 420 ; and/or 2) determining a filtered light sensor value based on the light intensity reading from light sensor 440 . One example of logic for determining a filtered infrared sensor value is module D' described with reference to the flowchart 2300 shown in FIG. 23 . According to some implementations, the control logic may determine a filtered infrared sensor value based on one or more sky sensors, one or more environmental sensors, or both sky sensors and environmental sensors. An example of logic for determining a filtered light sensor value is module C1' described with reference to flowchart 3100 shown in FIG .

In one instance, multi-sensor device 401 may execute instructions stored in memory for determining filtered sensor values and communicating the filtered sensor values to an external controller over a communication network. An example of an infrared cloud detector system 400 including a multi-sensor device that can communicate sensor readings and/or filtered values to an external controller via a communication network 1410 is shown in FIG. 14 . Control logic implemented by an external controller can make tinting decisions to determine a tint level and implement tint commands to shift the tint of one or more tintable windows in a building. This control logic is described with reference to modules A1 , B, C1 and D shown in FIGS. 22 , 24-28 and 30 .

- Instance B

32A and 32B show perspective views of a diagrammatic representation of an infrared cloud detector system 3200 including an infrared cloud detector in the form of a multi-sensor device 3201 and one or more external controllers (not shown) in communication with the multi-sensor device 3201 via a communication network (not shown), according to various implementations. 33A and 33B show perspective views of diagrammatic representations of internal components of a multi-sensor device 3301 according to one aspect. In one implementation, the multi-sensor device 3201 of FIGS. 32A and 32B may implement components of the multi-sensor device 3301 shown in FIGS. 33A and 33B .

In FIGS. 32A and 32B , the multi-sensor device 3201 includes a housing 3210 coupled to a mast 3220 . Mast 3220 may serve as a mounting assembly comprising a first end portion for coupling to base portion 3214 of housing 3210 and a second end portion for mounting to a building. In one example, the base portion 3214 is mechanically threaded therethrough or fixedly attached or otherwise coupled to or coupled with the first end portion of the mast 3220 via a rubber gasket. Mast 3220 may also include a second end portion, which may include a mounting or attachment mechanism for mounting or attaching mast 3220 to the roof of a building (e.g., on the roof of a building having room 3230 , as shown in FIG. 3 ), such as to a surface of the roof, a wall on the roof, or to another structure on the roof. The housing includes a cover 3211 formed of a light diffusing material. Cover 3211 also includes a thinned portion 3212 .

As shown in FIG. 32B , multi-sensor device 3201 also includes a first ambient temperature sensor 3222 located on the bottom exterior surface of base portion 3214 . The first ambient temperature sensor 3222 is configured to measure the ambient temperature of the external environment during operation. The first ambient temperature sensor 3222 is located on the bottom surface to help shield it from direct solar radiation, for example, when the infrared cloud detector system 3200 is located in an outdoor environment with the upper surface facing upward. The first ambient temperature sensor 3222 can be, for example, a thermistor, a thermocouple, a resistance thermometer, a silicon bandgap temperature sensor, and the like.

33A , 33B, and 33C show perspective views of diagrammatic representations of internal components of a multi-sensor device 3301 according to one aspect. The multi-sensor device 3301 generally includes a housing 3302 (partially shown) having a cover formed of light diffusing material including at least one thinned portion. The multi-sensor device 3301 also includes a diffuser 3304 . As shown, in some implementations, the housing 3302 and the diffuser 3304 are rotationally symmetric about an imaginary axis 3342 passing through the center of the multi-sensor device 3301 .

The multi-sensor device 3301 also includes a first infrared sensor device 3372 and a second infrared sensor device 3374 on an upper portion of the multi-sensor device 3301 and positioned behind a cover formed of a light diffusing material. Each of the first infrared sensor device 3372 and the second infrared sensor device 3374 includes an on-board infrared sensor for measuring sky temperature ( T _sky ) and an on-board ambient temperature sensor for measuring ambient temperature ( T _amb ). The first infrared sensor device 3372 is positioned to face outward from the upper surface of the multi-sensor device 3201 in a direction along the imaginary axis 3373 . The second infrared sensor device 3374 is positioned to face outward from the upper surface of the multi-sensor device 3201 in a direction along the imaginary axis 3375 . The multi-sensor device 3301 may also include an optional third infrared sensor device 3360 on an upper portion of the multi-sensor device 3301 and positioned behind a cover formed of a light diffusing material. The third infrared sensor device 3360 is a stand-alone infrared sensor, or includes both an on-board infrared sensor and an on-board ambient temperature sensor. Optional third infrared sensor device 3360 is positioned to face outward from the upper surface of multi-sensor device 3201 in a direction along imaginary axis 3361 . In the illustrated example, the first infrared sensor device 3372 and the second infrared sensor device 3374 are positioned such that their axes 3373 , 3375 face outward from the top portion of the housing (e.g., the housing shown in FIGS. 4A and 4B ) so that temperature readings based on infrared radiation captured from above the multi-sensor device 3301 can be taken during operation. In one aspect, the first infrared sensor device 3372 is separated from the second infrared sensor device 3374 by at least about one inch.

Optionally, the multi-sensor device 3301 may also include a free-standing ambient temperature sensor (not shown) located on the bottom exterior surface of the housing to shield it from direct solar radiation. Standalone ambient temperature sensors (eg, thermistors, thermocouples, resistance thermometers, silicon bandgap temperature sensors) are configured to measure the ambient temperature of the external environment during operation.

Returning to FIGS. 33A and 33B , the multi-sensor device 3301 includes a plurality of visible light sensors 3342 positioned behind a cover formed of a light diffusing material. While twelve (12) visible light photosensors 3342 are shown, it should be understood that a different number may be implemented. A plurality of visible light photosensors 3342 are positioned annularly along a ring (eg, the ring may have a center coincident with axis 3342 and may define a plane orthogonal to axis 3342 ). In this implementation, and more specifically, visible light photosensors 3342 may be positioned equidistantly along the circumference of the ring. Each of the visible light photosensors 3342 has a photosensitive region 3343 . The multi-sensor device 3301 also optionally includes an additional upward facing visible light sensor 3340 on the upper portion of the multi-sensor device 3301 . This optional VL light sensor sensor 3340 has an orientation axis parallel to and in some cases directed along and concentric with axis 3342 . The visible light sensor 3340 has a photosensitive area 3343 .

In some implementations, the viewing angle of each of the visible light sensors 3342 , 3340 is in the range of approximately 30 degrees to approximately 120 degrees. For example, in one particular application, the viewing angle is approximately 100 degrees. In some implementations, the distribution of incident light detectable by each of the visible light sensors 3342 , 3340 approaches a Gaussian (or "normal") distribution. Assuming that the light detected by each of the visible light photosensors 3342 , 3340 is associated with a Gaussian distribution, half of the power detected by each of the photosensors (the -3dB point) is found within a viewing cone defined by the viewing angle.

Diffuser 3304 is positioned around the perimeter of the ring of visible light sensor 3342 to diffuse light incident on the device before the light is sensed by light sensor 3342 . For example, diffuser 3304 can effectively act as a light integrator, which spreads or distributes incident light more evenly. This configuration reduces the likelihood that either visible light photosensor 3342 will receive a sufficient intensity of a tip reflection or glare (such as off a car windshield, metal surface, or mirror). The diffuser 3342 can also increase the detection of light incident at oblique angles. Figure 33A shows a diagrammatic representation of an example diffuser 3304 that can be used in the multi-sensor device 3301 of Figure 33A, according to some implementations. In some implementations, the diffuser 3304 is a single unitary structure having a ring shape. For example, diffuser 3304 may have a hollow cylindrical shape with an inner diameter, an outer diameter, and a thickness defined by the inner and outer diameters. In some implementations, the diffuser 3304 has a height that encompasses the field of view of each of the visible light photosensors 3342 (the field of view is defined by the viewing angle and the distance or spacing between the surface outside the photosensitive region of the visible light photosensor 3342 and the inner surface of the diffuser 3304 ).

During operation, the infrared sensors of the first infrared sensor device 3372 and the second infrared sensor device 3374 detect infrared radiation radiated from any object or medium within their field of view to measure the sky temperature ( T _sky ). The field of view is based on the physical and material properties of the infrared sensor. Based solely on their physical and material properties, some examples of infrared sensors have a field of view ranging from about 50 degrees to about 80 degrees. In one particular example, the infrared sensor has a field of view of about 70 degrees. During operation, the ambient temperature sensors of the first infrared sensor device 3372 and the second infrared sensor device 3374 measure the ambient temperature ( T _amb ). Although multi-sensor 3301 is shown with redundant infrared sensors, it should be understood that multi-sensor includes one or more infrared sensors. During operation, a plurality of visible light photosensors 3342 and upward facing photosensors 3340 positioned behind a cover formed of light diffusing material measure the intensity of the received visible light.

Returning to FIGS. 32A and 32B , the infrared cloud detector system 3200 also includes logic for making determinations based on sensor data from readings taken by the sensors of the multi-sensor device. In this case, the multi-sensor device and/or one or more external controllers (not shown) include memory and one or more processors that can execute instructions stored in memory (not shown) for using the logic of infrared cloud detector system 3200 .一或多個外部控制器與多感測器裝置通信(無線或有線)以接收具有由紅外線感測器(例如，第一紅外線感測器裝置3372及第二紅外線感測器裝置3374之紅外線感測器及/或紅外線感測器3360 )、環境溫度感測器(例如，第一紅外線感測器裝置3372及第二紅外線感測器裝置3374之環境溫度感測器或位於外殼之一底部外部表面上以將其對直射太陽輻射遮住之可選獨立式溫度感測器3222 )及可見光光感測器(例如，光感測器3342或面向上之光感測器3340 )取得之感測器讀數或經過濾感測器值的信號。 In some implementations, power/communication lines may run from a building or another structure to infrared cloud detector system 3200 . In one implementation, infrared cloud detector system 3200 includes a network interface that can be coupled to a suitable cable. The infrared cloud detector system 3200 can communicate data to one or more external controllers of the building via a network interface. In some other implementations, infrared cloud detector system 3200 may additionally or alternatively include a wireless network interface to enable wireless communication with one or more external controllers. In some implementations, infrared cloud detector system 3200 may also include a battery within or coupled to the housing to power the sensors and electrical components within it. Instead or in addition to power from a power supply (eg, from a building power supply), the battery can provide this power. In some implementations, infrared cloud detector system 3200 further includes at least one photovoltaic cell, eg, on a surface of the housing.

根據一個態樣，紅外線雲偵測器系統3200進一步包含用於將以下各者用作輸入來判定雲覆蓋條件之邏輯：當日時間、一年中之某天、來自紅外線感測器(例如，第一紅外線感測器裝置3372及第二紅外線感測器裝置3374之紅外線感測器及/或紅外線感測器3360 )中之一或多者之天空溫度讀數T _sky 、來自一或多個環境溫度感測器(例如，第一紅外線感測器裝置3372及第二紅外線感測器裝置3374之環境溫度感測器或位於外殼之一底部外部表面上以將其對直射太陽輻射遮住之可選獨立式溫度感測器3222 )之環境溫度讀數T _amb 及來自一或多個光感測器(例如，光感測器3342或面向上之光感測器3340 )之可見光強度讀數及來自紅外線感測器的溫度讀數T _sky 之振盪頻率。 Examples of this logic are described herein, for example, with respect to FIGS . 8-10 .

According to another aspect, the infrared cloud detector system 3200 further includes various logics described with reference to FIGS. 21 , 22 , 23 , 24 , 26 , 27 , 28 and 30 and 31 .在一個實施中，舉例而言，多感測器裝置3301及/或一或多個外部控制器包含用於以下操作之邏輯：1)基於來自紅外線感測器(例如，第一紅外線感測器裝置3372及第二紅外線感測器裝置3374之紅外線感測器及/或紅外線感測器3360 )中之一或多者之溫度讀數T _sky 及來自一或多個環境溫度感測器(例如，第一紅外線感測器裝置3372及第二紅外線感測器裝置3374之環境溫度感測器或位於外殼之一底部外部表面上以將其對直射太陽輻射遮住之可選獨立式溫度感測器3222 )之環境溫度讀數T _amb 判定一經過濾紅外線感測器值；及/或2)基於來自一或多個光感測器(例如，光感測器3342或面向上之光感測器3340 )之光強度讀數判定一經過濾光感測器值。 One example of logic for determining a filtered infrared sensor value is module D' described with reference to the flowchart 2300 shown in FIG. 23 . An example of logic for determining a filtered IR sensor value is module C1' described with reference to flowchart 3100 shown in FIG. 31 . In one instance, multi-sensor device 401 may execute instructions stored in memory for determining filtered sensor values and communicating the filtered sensor values to an external controller over a communication network. An example of an infrared cloud detector system including a multi-sensor device that can communicate sensor readings and/or filtered values to an external controller via a communication network 1410 is shown in FIG. 14 . Control logic implemented by an external controller can make tinting decisions to determine a tint level and implement tint commands to shift the tint of one or more tintable windows in a building. This control logic is described with reference to modules A1 , B, C1 and D shown in FIGS. 22 , 24-28 and 30 .

E. Comparison of Intensity Readings from Light Sensors and Delta Values During Different Cloud Conditions

As discussed above, infrared sensors can be more accurate than visible light sensors in detecting "clear" conditions in the early morning and evening. However, direct sunlight and other conditions can cause some noise that causes oscillations in the infrared sensor readings. If the frequency and/or amplitude of these oscillations are low, infrared sensor readings can be used to make a high confidence assessment of cloud cover conditions. Also, certain conditions (eg, fast moving clouds) can cause oscillations in light sensor readings. If the oscillation frequency is low, the light sensor readings can be used to make a high confidence assessment of cloud cover conditions during the day. In some implementations, logic may determine whether the oscillations in infrared sensor readings have a high frequency and/or whether the oscillations in light sensor readings have a high frequency. If it is determined that the oscillations in the infrared sensor readings have a high frequency, the logic uses the light sensor readings to determine cloud cover conditions. If it is determined that the oscillations in the light sensor readings have a high frequency, the logic uses the difference between the infrared sensor readings and the ambient temperature sensor readings to determine cloud cover conditions. To illustrate the technical advantage of using this logic to select the type of sensor reading that depends on the oscillation, FIGS. 5A , 5B , 6A , 6B , 7A and 7B contain graphs of intensity readings I taken by the visible light photosensor for comparison with the difference (difference ( Δ )) between the temperature reading T _sky taken by the infrared sensor and the temperature reading Tamb _taken by the ambient temperature sensor during different cloud cover conditions. The visible light sensor, infrared sensor, and ambient temperature sensor are similar to those sensors described with respect to the components of infrared cloud detector 310 shown in FIG. 3 . Each of the curves has readings taken during a time period of the day.

One advantage of implementing an infrared sensor in a multi-sensor device is that the amplitude of the oscillations will generally be lower as compared to a light sensor due to the generally larger field of view, light diffuser, and consistent response of the infrared sensor to heat throughout the day, so infrared based sensing can be made with higher confidence. Evaluation of line sensors.

5A - 5B contain graphs of plots of readings taken on a day with full sun and clear weather (only clouds passing by in the middle of the day). FIG. 5A is a graph with a plot 510 of intensity readings I taken by a visible light photosensor over time. 5B is a graph having a curve 520 of the difference (difference (Δ)) between the temperature reading T _sky taken by the infrared sensor and the temperature reading _Tamb taken by the ambient temperature sensor over time. As shown in curve 510 of FIG. 5A , the intensity reading I taken by the visible light photosensor is high most of the day and falls off in a high frequency (short period) oscillation when clouds pass in the middle of the day. Curve 520 of FIG. 5A shows that the value of Delta (Δ) did not increase above the lower threshold value during the day, which indicates a high confidence "clear" condition.

6A - 6B contain graphs of plots of readings taken on a day with frequent passing clouds in the morning until afternoon and later in the afternoon with two slow moving clouds passing by. FIG. 6A is a graph with a plot 610 of intensity readings I taken by a visible light photosensor over time. 6B is a graph having a curve 640 of the difference (Delta (Δ)) between the temperature reading T _sky taken by the infrared sensor over time and the temperature reading _Tamb taken by the ambient temperature sensor over time. As shown in curve 610 of FIG. 6A , the intensity reading I taken by the visible light sensor has a high frequency portion 620 during the time period of morning through afternoon when there are frequent cloud passages. Later in the afternoon when two slow moving clouds pass by, the curve 610 has a low frequency portion 630 . Curve 640 in FIG. 6B shows that the value of delta (Δ) has a high frequency during the time period with frequent cloud passing in the morning until the afternoon, and the value remains between the upper and lower thresholds, indicating intermittent cloudiness. The value of delta (Δ) later in the afternoon has a low frequency oscillation with values between the upper and lower thresholds, and the lower threshold is also shifted between "intermittently cloudy" and "clear" conditions. In this case, the infrared sensor values indicate high confidence "intermittent cloudy" conditions from the morning until the afternoon, and the light sensor values indicate high confidence "intermittent cloudy" conditions later in the afternoon.

7A - 7B contain graphs of plots of readings taken over time during a cloudy day (except for a brief period in the middle of the day). FIG. 7A is a graph with a plot 710 of intensity readings I taken by a visible light photosensor over time. 7B is a graph having a curve 720 of the difference (Delta (Δ)) between the temperature reading T _sky taken by the infrared sensor and the temperature reading _Tamb taken by the ambient temperature sensor over time. As shown in curve 710 of FIG. 7A , the intensity readings I taken by the visible light photosensor are low most of the day and increase in high frequency (short period) oscillations when the sky is clear for short periods of time in the middle of the day. Curve 720 of FIG. 7A shows that the value of delta (Δ) does not become lower than the upper threshold during the day, which indicates a high confidence "cloudy" condition.

In some implementations, the infrared cloud detector system uses the readings from the infrared sensors to assess the difference between the ambient temperature and the temperature readings from the infrared sensors that measure wavelengths in the infrared range (eg, wavelengths between 8 microns and 14 microns). In some cases, one or more correction factors are applied to the calculated delta difference. The delta difference provides a relative sky temperature value that can be used to classify cloud cover conditions. For example, a cloud cover condition may be determined to be in one of three buckets "clear," "cloudy," and "overcast." In the process of using this infrared cloud detector system, the cloud cover condition judged has nothing to do with whether the sun is out or whether it is before sunrise/sunset.

Infrared cloud detector systems according to certain implementations may have one or more technical advantages. For example, during early morning and evening conditions, an infrared sensor can determine whether it is cloudy or sunny independently of visible light intensity levels. This determination of cloud cover conditions during times when the sun is not out may provide additional context to determine the tint status (also referred to herein as "tint level") of tintable windows when the sensor would be ineffective during these times. As another example, an infrared sensor may be used to detect general cloud cover conditions within its field of view. This information can be used in conjunction with light sensor readings to determine whether "clear" or "cloudy" conditions as determined by the light sensor are likely to persist. For example, if the light sensor detects a sharp rise in intensity levels that would tend to indicate "clear" conditions, the infrared sensor indicates "cloudy" conditions, and the "clear" conditions are not expected to persist.

Conversely, if the infrared sensor reports a "clear" condition and the light sensor reading indicates it is a "clear" condition, then the "clear" condition is likely to persist. As another example, at the time the tintable window needs to be in steady state at sunrise, the transition needs to start X time before sunrise (eg, transition time). During this time, the light sensor is inefficient because there is minimal exposure. The IR sensor can determine pre-sunrise cloud conditions to inform control logic whether to start the tinting process (during clear skies), or to keep the tintable window clear when "cloudy" conditions at sunrise are expected.

III. Example of a method for determining cloud cover conditions using infrared and ambient temperature readings

8-10 show flowcharts describing methods of determining a cloud cover condition using readings from at least one infrared sensor and an ambient temperature sensor according to various embodiments. In FIGS. 9-10 , readings from at least one light sensor can also be used to determine cloud cover conditions under certain conditions. In some cases, the infrared sensor used to take the temperature reading is calibrated to detect infrared radiation in the spectrum of about 8 μm to 14 μm and/or has a field of view of about 72 degrees. In some cases, the light sensor used to take the light sensor reading is calibrated to detect the intensity of visible light (eg, between about 390 nm and about 700 nm) in the bright light range, which generally refers to light visible to the average human eye under good lighting conditions (eg, brightness levels ranging from about 10 cd/ ^m2 to about 108 cd/ ^m2 ). Although these methods are described with respect to readings from a single infrared sensor, a single ambient temperature sensor, and/or a single light sensor, it should be understood that values from multiple sensors of one type may be used, for example, multiple sensors oriented in different directions may be used. If multiple sensors are used, the method may use a single value based on a particular orientation of one sensor (eg, a functioning sensor), or take an average, mean, or other statistically related value of readings from multiple functioning sensors. In other cases, there may be redundant sensors, and the infrared cloud detector may have logic to use the value from a functioning sensor. For example, by evaluating which of the sensors is functioning and/or which is not based on comparing readings from various sensors.

A. Method I

FIG. 8 shows a flowchart 800 describing a method of determining cloud cover conditions using temperature readings from an infrared sensor and an ambient temperature sensor, according to an implementation. The infrared sensor and ambient temperature sensor of the infrared cloud detector system typically take readings periodically (at sample time). The processor executes the instructions stored in the memory to perform the operations of the method. In one implementation, the infrared cloud detector system has components similar to those described with respect to the system with the infrared cloud detector 100 described with respect to FIG. 1 . In another implementation, an infrared cloud detector system has components similar to those described with respect to the system with infrared cloud detector 310 in FIG. 3 .

In FIG. 8 , the method starts at operation 801 . At operation 810 , a signal is received at the processor having a sky temperature reading T _sky from the infrared sensor and _{an ambient temperature reading Tamb} from the ambient temperature sensor. Signals from the infrared sensor and/or the ambient temperature sensor are received wirelessly and/or via a wired electrical connection. The infrared sensor takes temperature readings based on the infrared radiation it receives within its field of view. Infrared sensors are typically oriented towards an area of the sky of interest, eg, an area above a building. The ambient temperature sensor is configured to be exposed to the external environment to measure the ambient temperature.

In operation 820 , the processor calculates the difference (difference (Δ)) between the temperature reading T _sky obtained by the infrared sensor and the temperature reading T _amb obtained by the ambient temperature sensor at the sample time. Optionally (represented by the dotted line), a correction factor is applied to the calculated delta (Δ) (operation 830 ). Some examples of applicable correction factors include humidity, sun angle/elevation, and site elevation.

In operation 840 , the processor determines whether the calculated delta (Δ) value is below a threshold value (eg, -5 millidegrees Celsius, -2 millidegrees Celsius, etc.). If it is determined that the calculated delta (Δ) value is below the lower threshold, then the cloud cover condition is determined to be a "clear" condition (operation 850 ). During operation of the infrared cloud detector, the method then increments to the next sample time and returns to operation 810 .

If it is determined that the calculated difference (Δ) is higher than the lower threshold, the processor determines in operation 860 whether the calculated difference (Δ) is higher than an upper threshold (eg, 0 millidegrees Celsius, 2 millidegrees Celsius, etc.). If it is determined in operation 860 that the calculated difference (Δ) is higher than the upper threshold, the processor determines the cloud coverage condition as a "cloudy" condition (operation 870 ). During operation of the infrared cloud detector, the method then increments to the next sample time and returns to operation 810 .

If it is determined in operation 860 that the calculated difference (Δ) is lower than the upper threshold, the processor determines the cloud coverage condition as "intermittent cloudy" or another intermediate condition (operation 880 ). During operation of the infrared cloud detector, the method then increments to the next sample time and returns to operation 810 .

B. Method II

9 shows a flowchart 900 depicting a method of determining cloud cover conditions using readings from an infrared sensor, an ambient temperature sensor, and a light sensor of an infrared cloud detector system, according to an implementation. Infrared sensors, ambient temperature sensors, and light sensors typically take readings periodically (at sample time). The infrared cloud detector system also includes a processor that executes instructions stored in the memory to perform the logical operations of the method. In one implementation, the infrared sensor, ambient temperature sensor, and light sensor are similar to the components of the infrared cloud detector system 300 described with respect to FIG. 3 . In another implementation, the infrared sensor, ambient temperature sensor, and light sensor are similar to the components of the infrared cloud detector system 400 described with respect to FIGS. 4A - 4C .

In FIG. 9 , the logic of the method begins at operation 901 . At operation 910 , one or more signals are received at the processor having a temperature reading T _sky from the infrared sensor at a _{particular sample time, a temperature reading Tamb} from the ambient temperature sensor at the sample time, and an intensity reading from the light sensor at the sample time. Signals from the infrared sensor, the ambient temperature sensor and the light sensor are received wirelessly and/or via a wired electrical connection. The infrared sensor takes temperature readings based on the infrared radiation it receives within its field of view. Infrared sensors are typically oriented towards an area of the sky of interest, eg, an area above a building. The ambient temperature sensor is configured to be exposed to the external environment to measure the ambient temperature. The sensing surface of the light sensor is also typically oriented towards the region of sky of interest, and direct sunlight is blocked or diffused from illuminating the sensing surface.

在操作920 ，所述邏輯判定當日時間是否在以下時間週期中之一者期間：(i)在日出前不久(例如，開始於日出前45分鐘、日出前30分鐘、日出前20分鐘或日出前之其他合適時間量的第一時間)且直至在日出稍後(例如，開始於日出後45分鐘、日出後30分鐘、日出後20分鐘或日出後之其他合適時間量的第二時間)之時間週期，及(iii)在日落前不久(黃昏)(例如，開始於日落前45分鐘、日落前30分鐘、日落前20分鐘或日落前之其他合適時間量的第三時間)且直至日落之時間週期。 In one instance, the time of sunrise can be determined from measurements taken by the visible wavelength light sensor. For example, time period (i) may end at the point when the visible light sensor starts measuring direct sunlight, ie, the point at which the intensity reading of the visible light sensor is at or above a minimum intensity value. Additionally or alternatively, time period (iii) may be determined to end at the point at which the intensity reading from the visible wavelength light sensor is at or below a minimum intensity value. In another example, the time of sunrise and/or the time of sunset can be calculated using a solar calculator and the day of the year, and the time periods (i) and (iii) can be calculated by a defined period of time (e.g., 45 minutes) around the calculated time of sunrise/sunset.

In some implementations, the logic determines whether the current time is during one of time periods (i), (ii), (iii) or (iv) based on a calculated sun elevation. The logic uses one of various publicly available codes to determine the sun's elevation for the current time. If it determines that the calculated solar elevation is less than 0, then the logic determines that the time is during the night time period (iv). If it is determined that the calculated solar elevation is greater than 0 and less than a first solar elevation threshold associated with a time immediately after sunrise (e.g., 10 minutes after sunrise, 20 minutes after sunrise, 45 minutes after sunrise, etc.), then the logic may determine that the time is within the following time period: (i) between immediately before sunrise and immediately after sunrise. In one example, the first sun elevation threshold is 5 degrees above the horizon. In another example, the first sun elevation threshold is 10 degrees above the horizon. If it is determined that the calculated solar elevation is less than 180 degrees and greater than a second threshold value associated with the time immediately before sunset (e.g., 10 minutes after sunset, 20 minutes after sunset, 45 minutes after sunset, etc.), then the logic may determine that the time is within the following time period: (iii) within the time period immediately before sunset fall between. In one example, the second sun elevation threshold is 175 degrees or 5 degrees from the horizon. In another example, the second sun elevation threshold is 170 degrees or 10 degrees from the horizon. If the logic determines that the calculated solar elevation is greater than the first solar elevation threshold and less than the second solar elevation threshold, then the logic may determine that the time is within the following time period: (ii) between time periods (i) and (iii).

If it is determined at operation 920 that the time of day is during either of the time periods (i) or (iii), the logic is implemented to calculate the difference (Δ) between the temperature reading T _sky taken by the infrared sensor and the temperature reading _Tamb taken by the ambient temperature sensor at a sample time (operation 930 ). Optionally (represented by the dotted line), a correction factor is applied to the calculated delta (Δ) (operation 930 ). Some examples of applicable correction factors include humidity, sun angle/elevation, and site elevation.

In one embodiment, the logic also determines at operation 920 whether the infrared readings oscillate at a frequency greater than a second defined level. If the processor determines at operation 920 that the time of day is within time period (i) or (iii) and the infrared readings oscillate at a frequency greater than a second defined level, the processor applies operation 990 to determine cloud conditions using the light sensor readings. For example, the processor may determine a "clear" condition if the light sensor reading is above a certain minimum intensity level, and a "cloudy" condition if the light sensor reading is at or below the minimum intensity level. If the system is still operating, the method advances to the next sample time and returns to operation 910 .

In operation 934 , the processor determines whether the calculated delta (Δ) value is below a threshold value (eg, -5 millidegrees Celsius, -2 millidegrees Celsius, etc.). If it is determined that the calculated delta (Δ) value is below the lower threshold value, then the cloud cover condition is determined to be a "clear" condition (operation 936 ). The method then increments to the next sample time and returns to operation 910 during operation of the infrared cloud detector.

If it is determined that the calculated difference (Δ) is higher than the lower threshold, the processor determines in operation 940 whether the calculated difference (Δ) is higher than an upper threshold (eg, 0 millidegrees Celsius, 2 millidegrees Celsius, etc.). If it is determined in operation 940 that the calculated difference (Δ) is higher than the upper threshold, the processor determines the cloud coverage condition as a "cloudy" condition (operation 942 ). If still in operation, the method advances to the next sample time and returns to operation 910 .

If it is determined in operation 940 that the calculated difference (Δ) is lower than the upper threshold, the processor determines the cloud coverage condition as "intermittently cloudy" or another intermediate condition (operation 950 ). If the system is still operating, the method advances to the next sample time and returns to operation 910 .

If it is determined at operation 920 that the time of day is not during any of time periods (i) or (iii), the processor determines that the time of day is during time period (ii) which is during the day after time period (i) and before time period (iii) (operation 960 ). If the processor determines at operation 960 that the time of day is during time period (ii) daytime, the processor calculates the difference between the temperature reading T _sky obtained by the infrared sensor and the intensity reading obtained by the light sensor (operation 970 ). At operation 980 , the processor determines that the calculated difference is within an acceptable limit. If the processor determines at operation 980 that the calculated difference is greater than the acceptable limit, the processor applies operation 930 to calculate the difference (Δ) and use the calculated difference (Δ) to determine the cloud cover condition, as discussed above.

In one embodiment, the processor also determines at operation 960 whether the infrared readings oscillate at a frequency greater than a second defined level. If the processor determines at operation 960 that the time of day is within time period (ii) and the infrared readings oscillate at a frequency greater than a second defined level, the processor applies operation 990 to determine cloud conditions using the light sensor readings. For example, the processor may determine a "clear" condition if the light sensor reading is above a certain minimum intensity level, and a "cloudy" condition if the light sensor reading is at or below the minimum intensity level. If the system is still operating, the method advances to the next sample time and returns to operation 910 .

If the processor determines at operation 980 that the calculated difference is within acceptable limits, then the light sensor readings are used to determine cloud cover conditions (operation 990 ). For example, the processor may determine a "clear" condition if the light sensor reading is above a certain minimum intensity level, and a "cloudy" condition if the light sensor reading is at or below the minimum intensity level. If the system is still operating, the method advances to the next sample time and returns to operation 910 .

In one embodiment, the processor also determines at operation 970 whether the light sensor readings oscillate with a frequency greater than a first defined level and the infrared readings oscillate with a frequency greater than a second defined level. If the processor determines at operation 980 that the calculated difference is within acceptable limits and the processor determines that the light sensor reading oscillates at a frequency greater than the first defined level, then the processor applies operation 930 to calculate the difference (Δ) and uses the calculated difference (Δ) to determine cloud cover conditions, as discussed above. If the processor determines at operation 980 that the calculated difference is not within acceptable limits and the processor determines that the infrared readings oscillate at a frequency greater than the second defined level, the processor applies operation 990 to determine cloud conditions using the light sensor readings. For example, the processor may determine a "clear" condition if the light sensor reading is above a certain minimum intensity level, and a "cloudy" condition if the light sensor reading is at or below the minimum intensity level. If the system is still operating, the method advances to the next sample time and returns to operation 910 .

In another embodiment, instead of or implementing operations 970, 980 , and 990 , the processor executes instructions to implement logic that executes both the daytime infrared sensor algorithm and the daylight sensor algorithm to independently determine cloudy/clear/intermediate conditions, each based on its own signal threshold and corresponding hue level. The control logic then applies the darker of the two tone levels determined independently by the daylight sensor algorithm and the daylight infrared sensor algorithm. An example of similar control logic is described with respect to operations 2820 , 2830 , 2832 , and 2840 depicted in FIG .

Returning to FIG. 9 , if the processor determines at operation 960 that the time of day is in the night time period (iv) after time period (iii) and before time period (i), the processor calculates a delta at operation 930 and uses the calculated delta (Δ) to determine cloud cover conditions, as discussed above.

IV. Method and system for controlling tintable windows using infrared sensor and/or light sensor readings

In an energy efficient building, the control logic used to set the level of its building systems may consider cloud cover in its decisions. For example, in a building with optically switchable windows (also referred to herein as "tintable windows"), the control logic may set the optical state of the optically switchable windows (e.g. For example, cloud cover is taken into account during the tint state of electrochromic windows. Conventional systems intended to provide this functionality typically use expensive sensing equipment to map the entire sky and track cloud movement. This mapping technique can suffer from not being able to align with the cloud until there is enough visible light to see it. Therefore, building systems may not need to be adjusted while aligning with the cloud.

In various implementations described herein, sensor data from an infrared cloud detector system (e.g., the system of FIG. 1 , system 300 in FIG. 3 , system 400 in FIGS. 4A-4C , or other infrared cloud detector systems described herein) can be used to classify building systems. As an example, this section describes control logic for determining cloud cover conditions and setting tint levels in one or more optically switchable windows (e.g., electrochromic windows) in a building based on the determined cloud cover conditions using readings (including infrared measurements) taken by sensors in an infrared cloud detector system. Although the control logic described in this section is described with reference to controlling the tint state in electrochromic windows, it should be understood that this logic can be used to control other types of optically switchable windows and other building systems. Electrochromic windows have one or more electrochromic devices, such as in U.S. Patent No. 8,764,950, issued July 1, 2014, and entitled "ELECTROCHROMIC DEVICES," and U.S. Patent Application No. 13/462,725, entitled "ELECTROCHROMIC DEVICES," filed May 2, 2012 (as U.S. Patent Application No. 9,261,751), each of which is hereby incorporated by reference in its entirety.

A. Electrochromic devices/windows

FIG. 10 schematically depicts an electrochromic device 1000 in cross section. The electrochromic device 1000 includes a substrate 1002 , a first conductive layer (CL) 1004 , an electrochromic layer (EC) 1006 , an ion conductive layer (IC) 1008 , a counter electrode layer (CE) 1010 and a second conductive layer (CL) 1014 . In one implementation, the electrochromic layer (EC) 1006 including tungsten oxide and the counter electrode layer (CE) 1010 includes nickel-tungsten oxide. Layers 1004 , 1006 , 1008 , 1010 , and 1014 are collectively referred to as an electrochromic stack 1020 . A voltage source 1016 operable to apply a potential across the electrochromic stack 1020 effects a transition of the electrochromic device, for example, between a bleached state (eg, as depicted in FIG. 11A ) and a colored state (eg, as depicted in FIG. 11B ). The order of the layers may be reversed with respect to the substrate 1002 .

In some cases, electrochromic devices have distinct layers and can be fabricated as all solid state devices and/or all inorganic devices. Examples of such devices and methods of making them are described in more detail in U.S. Patent Application Serial No. 12/645,111, entitled "Fabrication of Low-Defectivity Electrochromic Devices," and filed December 22, 2009 (issued as U.S. Patent No. 9,664,974), and entitled "Electrochromic Devices Devices) and U.S. Patent Application No. 12/645,159, filed December 22, 2009 (issued April 30, 2013 as U.S. Patent No. 8,432,603), both of which are hereby incorporated by reference in their entirety. It should be understood, however, that any one or more of the layers in the stack may contain some amount of organic material. It is useful for liquids that may be present in one or more layers in small amounts. It should also be understood that solid state materials may be deposited or otherwise formed by processes using liquid components, such as certain processes using sol-gels or chemical vapor deposition. In addition, it should be understood that references to transitions between bleached and colored states are non-limiting and suggest only one example among many of electrochromic transitions that may be implemented, and unless otherwise specified herein (including the foregoing discussion), whenever reference is made to a bleached-colored transition, the corresponding device or process encompasses other optical state transitions, such as non-reflective-reflective, transparent-opaque, etc. In addition, the term "bleached" refers to an optically neutral state, eg, colorless, transparent or translucent. Still further, unless otherwise specified herein, the "color" of an electrochromic transition is not limited to any particular wavelength or range of wavelengths. As understood by those skilled in the art, selection of appropriate electrochromic and counter electrode materials governs the associated optical transition.

In some implementations, an electrochromic device is configured to reversibly cycle between a bleached state and a colored state. When the electrochromic device is in the bleached state, a potential is applied to the electrochromic stack 1020 such that the available ions in the stack reside primarily in the opposing electrode 1010 . When the potential across the electrochromic stack is reversed, ions are transported across the ion conducting layer 1008 to the electrochromic material 1006 and transition the material to a colored state. In a similar manner, certain implementations of electrochromic devices described herein are configured to cycle reversibly between different shade levels (e.g., a bleached state, a darkest colored state, and intermediate levels between the bleached state and the darkest colored state).

Referring again to FIG. 10 , a voltage source 1016 is configured to operate in conjunction with the input from the sensor. As described herein, voltage source 1016 interfaces with a controller (not shown in this figure). Additionally, the voltage source 1016 may interface with an energy management system that controls the electrochromic device according to various criteria such as time of year, time of day, and measured environmental conditions. This energy management system combined with large area electrochromic windows can significantly reduce the energy consumption of buildings with said electrochromic windows.

Any material with suitable optical, electrical, thermal and mechanical properties may be used as the substrate 1002 or other substrates for the electrochromic stacks described herein. Examples of suitable substrates include, for example, glass, plastic, and mirror materials. Suitable glasses include clear or tinted soda lime glass, including soda lime float glass. Glass can be tempered or untempered. In many cases, the substrate is a glass sheet sized for residential window applications. The size of this glass pane can vary widely depending on the specific needs of the dwelling. In other cases, the substrate is architectural glass. Architectural glass is commonly used in commercial buildings, but can also be used in residential buildings, and often, but not necessarily, separates the indoor environment from the outdoor environment. In some examples, the architectural glass is at least 20 inches by 20 inches, and can be much larger, eg, as large as about 80 inches by 120 inches. Architectural glass is typically at least about 2mm thick, typically between about 3mm and about 6mm thick. Of course, the electrochromic device can be scaled smaller or larger with the substrate than the architectural glass. Additionally, electrochromic devices can be provided on mirrors of any size and shape.

Above the illustrated substrate 1002 is a conductive layer 1004 . In certain implementations, one or both of conductive layers 1004 and 1014 are inorganic and/or solid. Conductive layers 1004 and 1014 can be made from many different materials, including conductive oxides, thin metal coatings, conductive metal nitrides, and composite conductors. Typically, conductive layers 1004 and 1014 are transparent at least in the range of wavelengths at which electrochromism is exhibited by the electrochromic layers. Transparent conductive oxides include metal oxides and metal oxides doped with one or more metals. Examples of such metal oxides and doped metal oxides include indium oxide, indium tin oxide, doped indium oxide, tin oxide, doped tin oxide, zinc oxide, aluminum zinc oxide, doped zinc oxide, ruthenium oxide, doped ruthenium oxide, and the like. Because oxides are commonly used for these layers, they are sometimes referred to as "transparent conductive oxide" (TCO) layers. Thin metallic coatings that are substantially transparent may also be used, as well as combinations of TCO and metallic coatings.

The function of the conductive layer is to spread the potential provided by the voltage source 1016 on the surface of the electrochromic stack 1020 to the inner regions of the stack with relatively minimal ohmic potential drop. The potential is transferred to the conductive layer via the electrical connection to the conductive layer. In some aspects, bus bars (at least one in contact with conductive layer 1004 and at least one in contact with conductive layer 1014 ) provide electrical connection between voltage source 1016 and conductive layers 1004 and 1014 . The conductive layers 1004 and 1014 can also be connected to the voltage source 1016 by other conventional methods.

~0.3))內之插入使氧化鎢自透明(經漂白狀態)改變至藍(有色狀態)。 The conductive layer 1004 illustrated in the overlay is an electrochromic layer 1006 . In some aspects, electrochromic layer 1006 is inorganic and/or solid. The electrochromic layer may contain any one or more of a number of different electrochromic materials including metal oxides. Some examples of suitable metal oxides _include tungsten oxide ( _WO3 ), molybdenum oxide ₍ MoO3), niobium oxide ( _Nb2O5 ), titanium oxide (TiO2), copper _oxide ( _CuO ), iridium oxide ( _Ir2O3 ), chromium oxide ( _Cr2O3 ), manganese _oxide ( _{Mn2O3 )} , vanadium oxide ( _{V2O5 )} , nickel oxide ( _Ni2O3 ), cobalt oxide ( Co _{2 O 3} ) and the like. During operation, the electrochromic layer 1006 transfers ions to and receives ions from the counter electrode layer 1010 to cause a reversible optical transition . Typically, the coloration (or change of any optical property, such as absorbance, reflectance, and transmittance) of an electrochromic material is caused by reversible ion implantation (eg, intercalation) into the material and corresponding injection of charge-balancing electrons. Typically, a certain fraction of ions responsible for the optical transition is irreversibly bound in the electrochromic material. Some or all of the irreversibly bound ions are used to compensate for "blind charge" in the material. In most electrochromic materials, suitable ions include lithium ions (Li+) and hydrogen ions (H+) (ie, protons). However, in some cases other ions will be suitable. In various embodiments, lithium ions are used to generate electrochromism. Li-ion to tungsten oxide (WO _3-y (0<y

Insertion within ~0.3)) changes tungsten oxide from transparent (bleached state) to blue (colored state).

Referring again to FIG. 10 , in the electrochromic stack 1020 , the ion-conducting layer 1008 is sandwiched between the electrochromic layer 1006 and the opposing electrode layer 1010 . In some embodiments, the counter electrode layer 1010 is inorganic and/or solid. The counter electrode layer may comprise one or more of many different materials that act as a reservoir for ions when the electrochromic device is in the bleached state. During initiation of the electrochromic transition by, for example, applying an appropriate potential, the counter electrode layer transfers some or all of the ions it holds to the electrochromic layer, thereby changing the electrochromic layer to a colored state. Meanwhile, in the case of NiWO, the counter electrode layer is colored with the loss of ions. Suitable materials for the _counter electrode complementary to WO include nickel oxide (NiO), nickel tungsten oxide (NiWO), nickel vanadium oxide, nickel chromium oxide _, nickel aluminum oxide, nickel manganese oxide, nickel magnesium oxide, chromium oxide ( _Cr2O3 ), manganese oxide ( _MnO2 ), and Prussian blue. When the counter electrode 1010 made of nickel-tungsten oxide removes charge (ie, transports ions from the counter electrode 1010 to the electrochromic electrochromic 1006 ), the counter electrode layer 1010 will transition from a transparent state to a colored state.

In the illustrated electrochromic device 1100 , between the electrochromic layer 1006 and the counter electrode layer 1010 , there is an ion-conducting layer 1008 . The ion-conducting layer 1008 acts as a medium through which ions are transported (by way of electrolytes) when the electrochromic device transitions between the bleached state and the colored state. Preferably, the ion-conducting layer 1008 is highly conductive to the relevant ions for the electrochromic and counter electrode layers, but has sufficiently low electron conductivity that negligible electron transfer occurs during normal operation. Thin ion-conducting layers with high ion conductivity allow fast ion conduction and thus fast switching for high performance electrochromic devices. In certain aspects, ion conducting layer 1008 is inorganic and/or solid.

Examples of suitable materials for ion-conducting layers (ie, for electrochromic devices with distinct IC layers) include silicates, silicon oxides, tungsten oxides, tantalum oxides, niobium oxides, and borates. These materials can be doped with different dopants, including lithium. Lithium-doped silicon oxide includes lithium silicon-aluminum-oxide. In some embodiments, the ion-conducting layer includes a silicate-based structure. In one aspect, silicon aluminum oxide (SiAlO) is used for ion conducting layer 1008 .

In certain implementations, electrochromic device 1000 includes one or more additional layers (not shown), such as one or more passive layers. Passive layers to improve certain optical properties may be included in the electrochromic device 1000 . Passive layers for providing waterproof and scratch resistance may also be included in the electrochromic device 1000 . For example, the conductive layer may be treated with an anti-reflective or protective oxide or nitride layer. Other passive layers may be used to hermetically seal the electrochromic device 300 .

11A is a schematic cross-section of an electrochromic device in (or transitioning to) a bleached state. According to the illustrated example, an electrochromic device 1100 includes a tungsten oxide electrochromic layer (EC) 1106 and a nickel tungsten oxide counter electrode layer (CE) 1110 . The electrochromic device 1100 also includes a substrate 1102 , a conductive layer (CL) 11011 , an ion-conducting layer (IC) 1108 and a conductive layer (CL) 1114 . Layers 1104 , 1106 , 1108 , 1010 , and 1114 are collectively referred to as electrochromic stack 1120 . Power source 1116 is configured to apply a voltage potential and/or current to electrochromic stack 1120 via suitable electrical connections (eg, bus bars) to conductive layers 1104 and 1114 . In one aspect, the voltage source is configured to apply a potential of several volts to drive the transition of the device from one optical state to another. The polarity of the potential as shown in FIG. 11A is such that ions (in this example, lithium ions) reside primarily (as indicated by dashed arrows) in the nickel-tungsten oxide counter electrode layer 1110 .

FIG. 11B is a schematic cross-section of the electrochromic device 1100 shown in FIG. 11A but in a colored state (or transitioning to a colored state). In FIG. 11B , the polarity of the voltage source 1116 is reversed, making the tungsten oxide electrochromic layer 1106 more negative to accept additional lithium ions, and thereby transition to a colored state. Lithium ions are transported across the ion conducting layer 1108 to the tungsten oxide electrochromic layer 1106 as indicated by the dashed arrow. Tungsten oxide electrochromic layer 1106 is shown in or transitioning to a colored state. The nickel tungsten oxide counter electrode 1110 is also shown in or transitioning to a colored state. As explained, nickel tungsten oxide gradually becomes more opaque as it relinquishes (deintercalates) lithium ions. In this example, there is a synergistic effect in which the transition to the colored state for both layers 1106 and 1110 increases towards reducing the amount of light transmitted through the electrochromic stack and substrate.

In certain implementations, an electrochromic device includes an electrochromic (EC) electrode layer and a counter electrode (CE) layer separated by an ionically conductive (IC) layer that is highly conductive to ions and highly resistive to electrons. As is conventionally understood, the ionically conductive layer thus prevents shorting between the electrochromic layer and the counter electrode layer. The ionically conductive layer allows the electrochromic and opposing electrodes to hold charge and thereby maintain their bleached or colored state. In electrochromic devices having distinct layers, the components form a stack comprising an ion-conducting layer sandwiched between an electrochromic electrode layer and an opposing electrode layer. The boundaries between these three stacked components are defined by abrupt compositional and/or microstructural changes. Thus, the device has three distinct layers with two sharp interfaces.

According to some implementations, the counter electrode and the electrochromic electrode are formed in close proximity to each other, sometimes in direct contact, without separating to deposit an ionically conductive layer. In some implementations, electrochromic devices are used that have an interface region rather than a distinct IC layer. Such devices and methods of making them are described in US Patent No. 8,300,298, US Patent No. 8,582,193, US Patent No. 8,764,950, and US Patent No. 8,764,951, each of which is entitled "Electrochromic Devices" and each of which is hereby incorporated by reference in its entirety.

In certain implementations, an electrochromic device can be integrated into the insulating glass unit (IGU) of the electrochromic window, or can be in a single frame electrochromic window. For example, an electrochromic window may have an IGU that includes a first electrochromic lens and a second lens. The IGU also includes a spacer separating the first electrochromic lens from the second lens. The second lens in the IGU can be a non-electrochromic lens or others. For example, the second lens may have an electrochromic device thereon, and/or one or more coatings, such as low-E coatings and the like. Any of the lenses may also be laminated glass. Between the spacer and the first TCO layer of the electrochromic lens is a primary sealing material. This primary seal material is also Between the spacer and the second glass lens. Around the perimeter of the spacer is a secondary seal. These seals help keep moisture out of the interior space of the IGU. It also serves to prevent the escape of argon or other gases that may be introduced into the interior space of the IGU. The IGU also includes bus bar wiring for connection to the window controller. In some implementations, one or both of the bus bars are inside the finished IGU, however in one implementation, one bus bar is outside the seal of the IGU and one bus bar is inside the IGU. In the former embodiment, a region is used for sealing to one side of the spacer used to form the IGU. Thus, wires or other connections to the bus bars run between the spacers and the glass. Because many spacers are made of conductive metals, such as stainless steel, measures need to be taken to avoid short connections due to electrical communication between the bus bars and their connectors and the metal spacers.

B. Window Controller

The window controller is used to control the tint state (also referred to herein as "tint level") of one or more electrochromic devices in an electrochromic window or a zone of one or more electrochromic windows. In some embodiments, the window controller is capable of transitioning the electrochromic window between two tint states, a bleached state and a tinted state. In other embodiments, the controller may additionally toggle electrochromic windows (eg, windows with a single electrochromic device) between tint states including: a bleached state, one or more intermediate intermediate levels, and a tinted state. In some embodiments, the window controller is capable of transitioning the electrochromic window between four or more tint states. In other embodiments, the window controller is capable of switching an electrochromic window incorporating an electrochromic device between any number of shade levels between a bleached state and a colored state. Certain electrochromic windows allow for mid-tone gradations by using two (or more) electrochromic lenses in a single IGU, where each electrochromic lens is a two-state lens.

In some embodiments, an electrochromic window may comprise an electrochromic device on one lens of an insulating glass unit (IGU) and another electrochromic device on the other lens. If the window controller is able to switch each electrochromic device between two states (bleached state and colored state), then the IGU can achieve four different states (tint levels) - colored state where both electrochromic devices are colored state, a first intermediate state in which one electrochromic device is colored, a second intermediate state in which the other electrochromic device is colored, and a bleached state in which both electrochromic devices are bleached. Embodiments of multi-pane electrochromic windows, such as IGUs, are further described in U.S. Patent No. 8,270,059, entitled "MULTI-PANE ELECTROCHROMIC WINDOWS," to Robin Friedman et al. as the inventor, which is hereby incorporated by reference in its entirety.

In some embodiments, a window controller may be implemented to switch an electrochromic window having an electrochromic device capable of switching between two or more shade levels. For example, a window controller may be capable of transitioning electrochromic windows to a bleached state, one or more intermediate levels, and a tinted state. In some other embodiments, the window controller is capable of switching an electrochromic window incorporating an electrochromic device between any number of shade levels between a bleached state and a colored state. Embodiments of a method and controller for transitioning an electrochromic window to one or more midtone levels are further described in U.S. Patent No. 8,254,013, entitled "CONTROLLING TRANSITIONS IN OPTICALLY SWITCHABLE DEVICES," naming Disha Mehtani et al. International PCT application PCT/US17/35290 for CONTROL METHODS FOR TINTABLE WINDOWS IMPLEMENTING INTERMEDIATE TINT STATES", which is hereby incorporated by reference in its entirety.

In some embodiments, a window controller can power one or more electrochromic devices of an electrochromic window. Typically, this function of the window controller is augmented by one or more other functions described in more detail below. The window controllers described herein are not limited to window controllers that have the functionality to power their associated electrochromic devices for control purposes. That is, the power supply for the electrochromic window can be separate from the window controller, where the controller has its own power supply and directs power from the window power supply to the window. However, it is convenient to include a power supply with a window controller and configure the controller to power the window directly, as it eliminates the need for separate wiring for powering the electrochromic window.

In some cases, the window controller is a stand-alone controller configured to control the functions of a single window or multiple electrochromic windows without integration of the window controller into a building control network or a building management system (BMS). However, window controllers can be integrated into a building control network or a BMS, as further described in this section.

12 depicts a block diagram of components of a window controller 1250 and components of a window controller system according to implementations discussed herein.圖12為窗控制器1250之簡化方塊圖，且關於窗控制器之更多細節可見於美國專利申請案13/449,248及13/449,251中，兩者皆指明Stephen Brown為發明人，題目皆為《用於光學可切換窗之控制器(CONTROLLER FOR OPTICALLY-SWITCHABLE WINDOWS)》且兩者皆在2012年4月17日提交，且可見於美國專利申請案13/449,235(作為美國專利第8,705,162號發佈)，其題目為《控制光學可切換裝置中之轉變(CONTROLLING TRANSITIONS IN OPTICALLY SWITCHABLE DEVICES)》，指明Stephen Brown等人為發明人，且在2012年4月17日提交；所有所述申請案在此被以引用的方式全部併入。

In FIG. 12 , the illustrated components of window controller 1250 include a microprocessor 1255 or other processor, a pulse width modulator 1260 (PWM), a signal conditioning module 1265 , and a computer readable medium (e.g., memory) 1270 having a configuration file 1275 . Window controller 1250 is in electronic communication (wired or wireless) with one or more electrochromic devices 1200 in the electrochromic window via network 1280 to send control commands to one or more electrochromic devices 1200 . In some embodiments, window controller 1250 may be a local window controller that communicates (wired or wirelessly) with a host window controller over a network. In other examples, the window controller omits a signal conditioning module.

In some examples discussed herein, a building has one or more electrochromic windows between the exterior and interior of the building, and one or more sensors (e.g., light sensors, infrared sensors, ambient temperature sensors, etc.) located on the exterior of the building and/or inside one or more rooms with electrochromic windows. Outputs from one or more sensors may be received as inputs to signal conditioning module 1265 of window controller 1250 (eg, via a communications network). In some cases, output from one or more sensors may be received as input to a building management system (BMS), as described further in this section. Although the sensors of the depicted embodiments are shown as being located on a roof, the sensors may additionally or alternatively be located in other locations, such as outside a vertical wall of a building, inside a room, or on other surfaces to the outside. In some examples, a multi-sensor device having multiple sensors is located within or proximate to a housing. In some of these examples, two or more sensors in a multi-sensor arrangement are operable to measure the same or nearly the same data (e.g., two infrared sensors directed to the same general area of the sky), which can provide redundancy in case one of the sensors fails or has an otherwise erroneous reading.

An exterior light sensor to the building (e.g., one of a multi-sensor installation on a roof) can be used to detect radiant light incident on the light sensor, either directly from a light source such as the sun, or from light reflected to the sensor from a surface, particles in the atmosphere, clouds, etc. Each photosensor can generate a signal in the form of a current generated from the photoelectric effect, and the signal is a function of light incident on the photosensor. In some cases, light sensors detect radiant light in terms of irradiance in watts per square meter or other similar units. In other cases, light sensors detect light in the visible range of wavelengths in foot-candles or similar units. In many cases there is a linear relationship between these values of irradiance and visible light.

Irradiance values from sunlight during solar conditions can be predicted based on the time of day and time of year as the angle at which the sun hits the Earth varies. An external light sensor can detect actual radiant light in real time, taking into account reflected and blocked light due to buildings or other structures, changes in weather (eg, clouds), and the like. For example, on a cloudy day, sunlight may be blocked by clouds and the radiant light detected by the external sensor will be lower than on a cloudless (sun) day.

In the morning and evening, the sunlight level is low and the corresponding readings taken by the external light sensor are low values, which can also be considered consistent with readings taken during cloudy conditions during the day. out of For this reason, exterior light sensor readings taken during morning and evening may falsely indicate a cloudy condition (if considered separately). Additionally, any obstruction from buildings or mountains/mountains could also result in a false positive indication for cloudy days based on the exterior light sensor readings alone. Furthermore, using the external light sensor value alone would result in false positives for cloudy conditions immediately before sunrise, which could result in switching the electrochromic window to a transparent state at sunrise, allowing for glare conditions in rooms with clear windows.

In some embodiments, readings taken by at least two infrared sensors can be used to determine cloud conditions at times immediately before sunrise and also in the morning and at night. These infrared sensors can be operated independently of sunlight levels, allowing the tinting control logic to determine cloud conditions before sunrise, and as the sun sets, determine and maintain the proper tint state for the electrochromic windows during the morning and evening periods. Additionally, at least two infrared sensors can be used to detect cloud conditions even when the sensors are covered or otherwise blocked from direct sunlight.

In some aspects, a single device (sometimes referred to herein as an "infrared sensor device,""infrared cloud detector," or "multi-sensor device") includes an infrared sensor for detecting thermal radiation and an onboard ambient temperature sensor. Infrared sensors are typically positioned to be directed into the sky to measure the sky temperature ( T _sky ). On-board ambient temperature sensors are typically positioned to measure the ambient temperature ( T _amb ) at the device. Additionally or alternatively, the infrared sensor means outputs a temperature reading - delta (Δ) - the difference between the sky temperature reading and the ambient temperature reading. The infrared sensor device temperature readings ( T _sky , Tamb and/or Δ ) are usually _in degrees, eg, millidegrees Celsius or millidegrees Fahrenheit.

According to certain aspects, there may be multiple sensors associated with a single electrochromic window of a building or multiple electrochromic windows (eg, zones of electrochromic windows) of a building. For example, the multiple sensors may be in the form of a multi-sensor device having at least two infrared sensors, an ambient temperature sensor (eg, part of an infrared sensor), and multiple light sensors. For example, a multi-sensor device may be located on the roof of a building with one or more electrochromic windows. In one embodiment, outputs from redundant sensors are compared to each other to determine, for example, whether one of the sensors is controlled by a Obscuration by objects, such as a bird falling on a multi-sensor device on a roof. In some cases, it may be desirable to use relatively few sensors in a building because having many sensors may be expensive and/or some sensors may be unreliable. In some implementations, a single sensor or relatively few sensors (eg, 2, 3, 4, 5) may be used to determine the current level of radiant light from sunlight striking a building or possible sides of a building. Clouds may pass in front of the sun, or a construction vehicle may be parked in front of the setting sun. This occurrence will result in a deviation from the amount of radiant light from the sun that would be calculated to fall on the building normally during clear sky conditions.

In examples with light sensors, the light sensors may be, for example, charge coupled devices (CCDs), photodiodes, photoresistors, or photovoltaic cells. Those of ordinary skill in the art will understand that future developments in light sensors and other sensor technologies will also apply as they measure light intensity and provide an electrical output indicative of light levels.

In some embodiments, the output from the sensor may be input to the signal conditioning module 1265 . The input may be in the form of a voltage signal to the signal conditioning module 1265 . The signal conditioning module 1265 passes the output signal to the microprocessor 1255 or other processors. The microprocessor 1255 or other processor determines the tint level of the electrochromic window based on the information from the configuration file 1275 and based on the output or modification value from the signal conditioning module 1265 . Microprocessor 1255 then sends instructions to 1260 to apply voltage and/or current to electrochromic device 1200 of one or more electrochromic windows of a building via network 1280 to transition the electrochromic windows to a desired tint level.

In one aspect, the signal conditioning module 1265 is part of a multi-sensor device (eg, a rooftop multi-sensor device) that receives an output from one or more sensors of the multi-sensor device. In this case, the signal conditioning module 1265 communicates the output signal to the microprocessor 1255 or other processors of the window controller 1250 via a wired or wireless network. Microprocessor 1255 or other processor determines the tint level of the electrochromic window and sends instructions to PWM 1260 to apply voltage and/or current to electrochromic device 1200 of one or more electrochromic windows of a building via network 1280 to transition the electrochromic window to said desired tint level.

In some embodiments, microprocessor 1260 may direct PWM 1260 to apply voltage and/or current to the electrochromic window to transition it to any of four or more different hue levels. In one instance, the electrochromic window can be switched to at least eight different shade levels described as: 0 (brightest), 5, 10, 15, 20, 25, 30, and 35 (darkest). The tint level may correspond linearly to the visual transmittance value and solar heat gain coefficient (SHGC) value of light transmitted through the electrochromic window. For example, using the above eight hue grades, the brightest hue grade 0 may correspond to a SHGC value of 0.80, hue grade 5 may correspond to a SHGC value of 0.70, hue grade 10 may correspond to a SHGC value of 0.60, hue grade 15 may correspond to a SHGC value of 0.50, hue grade 20 may correspond to a SHGC value of 0.40, hue grade 25 may correspond to a SHGC value of 0.30, hue grade 30 may correspond to an SHGC value of 0.20, and hue grade 35 ( darkest) may correspond to a SHGC value of 0.10.

Window controller 1250 or a master controller in communication with window controller 1250 may use any one or more control logic components to determine a desired tint level based on signals from sensors and/or other inputs. Window controller 1250 may instruct PWM 1260 to apply voltage and/or current to one or more electrochromic windows of electrochromic device 1200 to transition them to a desired tint level.

C. Building Management System (BMS)

The window controllers described herein are also suitable for integration with a building management system (BMS). BMS is a computer-based control system installed in a building, which monitors and controls the mechanical and electrical equipment of the building, such as ventilation, lighting, electrical systems, elevators, fire protection systems and security systems. A BMS consists of hardware (including interconnections through communication channels to one or more computers) and associated software for maintaining conditions in a building according to preferences set by occupants and/or by the building manager. For example, a BMS can be implemented using a local area network such as Ethernet. Software may be based on, for example, Internet protocols and/or open standards. One example is software from Tridium, Inc. (Richmond, VA). One communication protocol commonly used by BMS is BACnet (Building Automation and Control network).

BMSs are most common in large buildings and typically at least function to control the environment within the building. For example, a BMS can control the temperature, carbon dioxide level and humidity inside a building. Typically, there are many mechanical devices controlled by the BMS, such as heaters, air conditioners, blowers, vents, and the like. To control the building environment, the BMS can switch these various devices on and off under defined conditions. One of the core functions of a typical modern BMS is to maintain a comfortable environment for the building's occupants while minimizing heating and cooling costs/demand. Therefore, a modern BMS is not only used for monitoring and control, but also to optimize the synergy between various systems, for example, to conserve energy and reduce building operating costs.

In some embodiments, a window controller is integrated with the BMS, where the window controller is configured to control one or more electrochromic windows or other tintable windows. In one embodiment, one or more electrochromic windows comprise at least one all-solid-state and inorganic electrochromic device, but may comprise more than one electrochromic device, eg, where each lens or bezel of the IGU is tintable. In one embodiment, one or more electrochromic windows include only all solid state and inorganic electrochromic devices. In one embodiment, the electrochromic window is a multi-state electrochromic window, as described in U.S. Patent Application Serial No. 12/851,514 (now U.S. Patent No. 8,705,162), filed August 5, 2010, and entitled "Multipane Electrochromic Windows," which is hereby incorporated by reference in its entirety.

13 depicts a schematic diagram of one embodiment of a BMS 1300 that manages many systems of a building 1301 , including security systems, heating/ventilating/air conditioning (HVAC), building lighting, electrical systems, elevators, fire protection systems, and the like. Security systems may include magnetic card access, turnstiles, solenoid actuated door locks, surveillance cameras, burglar alarms, metal detectors, and the like. Fire protection systems may include fire alarm and fire suppression systems (including water piping controls). The lighting system may include interior lighting, exterior lighting, emergency warning lights, emergency exit signs and emergency floor exit lighting. The power system may include primary generators, backup generators, and an uninterruptible power supply (UPS) grid.

Also, the BMS 1300 manages the main window controller 1302 . In this example, master window controller 3102 is depicted as a distributed network of window controllers including a master grid controller 1303 , intermediate grid controllers 1305a and 1305b , and end or leaf controllers 1310 . End or leaf controller 1310 may be similar to window controller 1250 described with respect to FIG. 12 . In one example, the main network controller 1303 may be closest to the BMS 1300 , and each floor of the building 1301 may have one or more intermediate network controllers 1305a and 1305b , while the windows of the building have their own end or leaf controllers 1310 . In this example, each of the end or leaf controllers 1310 controls a specific electrochromic window of the building 1301 .

Each of the tip or leaf controllers 1310 may be in a separate location from the electrochromic window it controls, or may be integrated into the electrochromic window. For simplicity, only ten electrochromic windows of building 1301 are depicted as being controlled by master window controller 1302 . In a typical setup, there may be a large number of electrochromic windows in a building controlled by master window controller 1302 . In such implementations, master window controller 1302 need not be a distributed network of window controllers. For example, the master window controller 1302, which is a single-ended controller controlling the function of a single electrochromic window, is also within the scope of the embodiments disclosed herein as described above.

One aspect of the disclosed embodiments is a BMS that includes a multi-sensor device (eg, multi-sensor device 401 shown in FIGS. 4A - 4C ) or other form of infrared cloud detector system. By incorporating feedback from the infrared cloud detector system, the BMS can provide, for example, enhanced: 1) environmental control, 2) energy savings, 3) security, 4) flexibility in control options, 5) improved reliability and service life of other systems due to less reliance on them and thus less maintenance, 6) information availability and diagnostics, and 7) efficient use of and higher productivity from workers, and various combinations of these due to automatic control of electrochromic windows. In some embodiments, the BMS may not be present, or the BMS may be present but may not communicate with the master network controller, or communicate with the master network controller at a high level. In certain embodiments, maintenance of the BMS will not interrupt control of the electrochromic window.

In some cases, the system of BMS 1300 may operate according to a daily, monthly, quarterly or yearly schedule. For example, lighting control systems, window control systems, HVAC, and security systems may operate on a 24-hour schedule, which takes into account the time people are in the building during the workday. At night, the building can enter an energy saving mode, and during the day, the system can operate in a manner that minimizes the building's energy consumption while providing occupant comfort. As another example, the system may shut down or enter an energy saving mode during the holiday period.

Scheduling information can be combined with geographic information. Geographical information may include the latitude and longitude of buildings. Geographical information may also include information about the direction each side of the building is facing. Using this information, different rooms on different sides of the building can be controlled differently. For example, in winter, for an east-facing room of a building, the window controller may direct the windows to have no tint in the morning, so that due to sunlight shining in the room, the room warms, and the lighting control panel may direct the lights to dim due to the illumination from the sun. West facing windows can be controlled by the occupants of the room in the morning, since the tint of windows on the west side can have no effect on energy savings. However, the mode of operation of the east-facing and west-facing windows can be switched at night (eg, when the sun goes down, the west-facing windows are untinted to allow sunlight to bring heat and illumination).

Additionally, the temperature within a building can be affected by external light and/or external temperature. For example, on a cold day and where the building is heated by a heating system, rooms closer to doors and/or windows will lose heat faster than, and be cooler than, interior areas of the building.

For implementations with external sensors, the building may include external sensors on the roof of the building. Alternatively, the building may include an exterior sensor associated with each exterior window or an exterior sensor on each side of the building. Exterior sensors on each side of the building can track the irradiance on one side of the building as the sun changes position during the day.

An example of a building, such as building 1301 in FIG. 13 , includes a building network or BMS, tintable windows for building exterior windows (ie, windows that separate the interior of the building from the exterior of the building), and many different sensors. Light from an exterior window of a building typically has an effect on interior lighting in the building from the window to about 20 or about 30 feet. That is, spaces in a building that are greater than about 20 feet or about 30 feet from an exterior window receive very little light from the exterior window. These spaces in the building away from the exterior windows are primarily illuminated by the building's interior lighting system.

14 is a block diagram of components of a system 1400 for controlling the function (eg, switching to different tint levels) of one or more tintable windows of a building (eg, building 1301 shown in FIG. 13 ), according to an embodiment. System 1400 may be one of the systems managed by a BMS (eg, BMS 1300 shown in FIG. 13 ), or may operate independently of the BMS.

System 1400 includes a master window controller 1402 that can send control signals to one or more tintable windows to control their functions. System 1400 also includes a network 1410 in electronic communication with master window controller 1402 . Control logic and instructions and/or sensor data for controlling the functions of the tintable windows may be communicated to the main window controller 1402 via the network 1410 . The network 1410 can be a wired or wireless network (eg, cloud network). In one embodiment, the network 1410 can communicate with the BMS to allow the BMS to send instructions for controlling the tintable windows over the network 1410 to the tintable windows in the building.

System 1400 also includes an EC device 400 in each of one or more tintable windows (not shown), and optional wall switch 1490 , all in electronic communication with primary window controller 1402 . In this illustrated example, master window controller 1402 may send a control signal to EC device 1401 to control the tint level of the tintable window with EC device 400 . Each wall switch 1490 also communicates with the EC device 1401 and the main window controller 1402 . An end user (eg, an occupant of a room with a tintable window) can use the wall switch 1490 to control the tint level and other functions of the tintable window with the associated EC device 1401 .

In FIG. 14 , the master window controller 1402 is depicted as a distributed network of window controllers comprising a master grid controller 1403 , a plurality of intermediate grid controllers 1405 in communication with the master grid controller 1403 , and a plurality of end or shutter controllers 1410 . Each of the plurality of end or shutter controllers 1410 communicates with a single intermediate network controller 1405 . Although master window controller 1402 is illustrated as a distributed network of window controllers, in other implementations master window controller 1402 can be a single window controller that controls the functions of a single tintable window. The components of system 1400 in FIG. 14 may be similar in some respects to those described with respect to FIG. 13 . For example, primary network controller 1403 can be similar to primary network controller 1303 , and intermediate network controller 1405 can be similar to intermediate network controller 1305 . Each of the window controllers in the distributed network of FIG. 14 includes a processor (eg, microprocessor) and a computer-readable medium in electrical communication with the processor.

In Figure 14 , each lobe or end window controller 1410 communicates with the EC device 1401 of a single tintable window to control the tint level of that tintable window in the building. In the case of an IGU, a leaf or end window controller 1410 may communicate with EC devices 1401 on multiple lenses of the IGU to control the tint level of the IGU. In other embodiments, each leaf or end window controller 1410 may communicate with multiple tintable windows, such as a zone of tintable windows. The leaf or end window controller 1410 may be integrated into the tintable window, or may be separate from the tintable window it controls. The leaf and end window controller 1410 in FIG. 14 may be similar to the end leaf controller 1410 in FIG. 13 , and/or may also be similar to the window controller 1250 described with respect to FIG. 12 .

Each wall switch 1490 can be operated by an end user (eg, the occupant of the room) to control the tint level and other functions of the tintable windows in communication with the wall switch 1490 . The end user can operate the wall switch 1490 to communicate a control signal to the EC device 400 in the associated tintable window. In some cases, these signals from wall switch 1490 may alter the signal from master window controller 1402 . In other situations (eg, high demand situations), the control signal from the main window controller 1402 may alter the control signal from the wall switch 1490 . Each wall switch 1490 also communicates with the leaf or end window controller 1410 to send information (e.g., time, date, requested tint level, etc.) about the control signal sent from the wall switch 1490 back to the main window controller 1402 , e.g., for storage in memory. In some cases, wall switch 1490 may be manually operated. In other cases, the wall switch 1490 may be controlled wirelessly by the end user using a remote device (e.g., cellular phone, tablet, etc.) that communicates wirelessly via control signals, e.g., using infrared (IR) and/or radio frequency (RF) signals. In such cases, the wall switch 1490 may include a wireless protocol chip, such as Bluetooth, EnOcean, WiFi, Zigbee, and the like. While the wall switch 1490 depicted in FIG. 14 is located on a wall, other embodiments of the system 1400 may have switches located elsewhere in the room.

System 1400 also includes a multi-sensor device 1412 in electronic communication with one or more controllers via communications network 1410 for communicating sensor readings and/or filtered sensor values to the controllers.

D. Logic for Controlling Electrochromic Devices/Windows

In some implementations, a controller (e.g., local end or louver controller, master or intermediate network controller, master window controller, etc.) includes intelligent control logic for calculating, determining, selecting, or otherwise generating the tint state for one or more optically switchable windows (e.g., electrochromic windows) of a building. This control logic can be used to determine a cloud cover condition based on sensor data from an infrared cloud detector system at the building, and use the determined cloud cover condition to determine the tint state of the switchable window. Such control logic may be used to implement methods for determining and controlling a desired tint level for one or more electrochromic windows or other tintable windows to account for occupant comfort, energy conservation, and/or other considerations. In some cases, the control logic uses one or more logic modules.

For example, Figures 15A - 15C depict general inputs to each of three logic modules A, B, and C of an exemplary control logic, according to certain implementations. Other examples of modules A, B, and C are described in International Patent Application PCT/US16/41344, entitled "CONTROL METHOD FOR TINTABLE WINDOWS", filed on July 7, 2016, and PCT/US, filed on May 5, 2015, entitled "CONTROL METHOD FOR TINTABLE WINDOWS"15/29675; each of said applications is hereby incorporated by reference in its entirety. Other examples of logic modules are described in International Patent Application PCT/US17 PCT/US17/66198, entitled CONTROL METHOD FOR TINTABLE WINDOWS, filed December 13, 2017, which is hereby incorporated by reference in its entirety. Another example of illustrative control logic involving four (4) modules is described later in this section.

- Examples of modules A, B and C

15A - 15C include diagrams depicting certain general inputs to each of the three logic modules A, B, and C of an exemplary control logic of the disclosed implementation. Each figure depicts a schematic side view of a room 1500 of a building with a table 1501 and an electrochromic window 1505 located between the exterior and interior of the building. The figure also depicts an infrared cloud detector system 1502 according to one example. In other implementations, other examples of the infrared cloud detector systems described herein can be used. In the illustrated example, infrared cloud detector system 1502 includes an infrared cloud detector 1530 located on the roof of a building. The infrared cloud detector 1530 has a housing 1532 with a cover made of light diffusing material, an infrared sensor 1534 and a light sensor 1510 within the housing of the housing 1532 , and an ambient temperature sensor 1536 located on a covered surface of the housing 1532 . Infrared sensor 1534 is configured to take a temperature reading T _sky based on infrared radiation received from the region of sky within its cone-shaped field of view. The ambient temperature sensor 1536 is configured to take an ambient temperature reading T _amb of the ambient air surrounding the infrared cloud detector 1530 . In one aspect, the infrared sensor and the ambient temperature sensor are integrated into the same sensor. The infrared sensor 1534 includes an imaginary axis that is perpendicular to the sensing surface of the infrared sensor 1534 and passes through its center. Infrared sensor 1534 is directed such that its sensing surface faces upward and can receive infrared radiation from the region of the sky within its field of view. The ambient temperature sensor 1536 is located on a covered surface to prevent direct sunlight from hitting its sensing surface. Although not shown, infrared cloud detector 1530 also includes one or more structures that hold its components within housing 1532 .

The infrared cloud detector system 1502 also includes a local window controller 1550 having a processor (not shown) that can execute instructions stored in memory (not shown) for implementing control logic to control the tint level of the electrochromic window 1505 . The local window controller 1550 communicates with the electrochromic window 1505 to send control signals. The local window controller 1550 is also in communication (wireless or wired) with the infrared sensor 1534 and the ambient temperature sensor 1536 to receive signals with temperature readings. The local window controller 1550 also communicates (wirelessly or wired) with the light sensor 1510 to receive a signal with a visible light intensity reading.

According to some aspects, power/communication lines run from a building or another structure to infrared cloud detector 1530 . In one implementation, the infrared cloud detector 1530 includes a network interface that can couple the infrared cloud detector 1530 to a suitable cable. The infrared cloud detector 1530 can communicate data to the local window controller 1550 or another controller of the building (eg, a network controller and/or a master controller) via a network interface. In some other implementations, infrared cloud detector 1530 may additionally or alternatively include a wireless network interface to enable wireless communication with one or more external controllers. In some aspects, infrared cloud detector 1530 may also include a battery within or coupled to its housing to power the sensors and electrical components within it. Instead or in addition to power from a power supply (eg, from a building power supply), the battery can provide this power. In one aspect, infrared cloud detector 1530 further includes at least one photovoltaic cell, eg, on one of the outer surfaces of the housing. The at least one photovoltaic cell may provide power instead or in addition to power provided by any other power supply.

FIG. 15A shows the penetration depth of direct sunlight into the room 1500 through the electrochromic window 1505 between the exterior and interior of the building containing the room 1500 . Penetration depth is measured as one of how far direct sunlight can penetrate into a room 1500 . As shown, penetration depth is measured in the horizontal direction away from the sill (bottom) of window 1505 . Typically, the window defines an aperture that provides an acceptance angle for direct sunlight. Penetration depth is calculated based on the geometry of the window (e.g., window size), its location and orientation in the room, any fins or other external shading outside the window, and the position of the sun (e.g., the angle of direct sunlight at a particular time of day and on a particular day). External shading to the electrochromic window 1505 may be due to any type of structure that can shade the window, such as overhangs, fins, etc. In FIG. 15A , there is an overhang 1520 above the electrochromic window 1505 , which blocks a portion of direct sunlight from entering the room 1500 , thus shortening the penetration depth.

Module A may be used to determine tint levels that take into account occupant comfort from direct sunlight passing through the electrochromic window 1505 onto the occupant or their active area (also referred to herein as "glare conditions"). The tint level is determined based on the calculated penetration depth of direct sunlight into the room and the type of space in the room at a particular moment (time of day and day of year) (eg, table near a window, hallway, etc.). In some cases, the tint scale may also be based on providing sufficient natural lighting into the room. In some cases, the depth of penetration is calculated at a future time to account for the vitrification time (the time required for the window to tint, eg, to 80%, 90%, or 100% of the desired shade level). The problem addressed in Module A is that direct sunlight can penetrate so deep into room 1500 that it falls directly on occupants working at a desk or other work surface in the room. The position of the sun can be calculated using publicly available programs and allows easy calculation of penetration depth.

15A to 15C also show desk 1501 in room 1500 as an example of a single occupancy office space type with desks associated with active areas (ie, desks) and locations of active areas (ie, desk locations). Each space type is associated with a different tone level for occupant comfort. For example, if the activity is a key activity such as work in an office being done at a desk or computer, and the desk is located near a window, then the desired tint level may be higher than if the desk is further away from the window. As another example, if the activity is non-critical, such as in the foyer, the desired tone level may be lower than in the same space with a desk for an office.

15B shows direct sunlight and radiation entering room 1500 through electrochromic window 1505 under clear sky conditions, according to an implementation. Radiation can come from sunlight scattered by molecules and particles in the atmosphere. Module B determines a tint level based on the calculated irradiance under clear sky conditions flowing through the electrochromic window 1505 under consideration. Various software such as the open source RADIANCE program can be used to calculate clear sky irradiance at a latitude, longitude, time of year, and time of day, and for a given window orientation.

15C shows radiant light from the sky as may be blocked by or reflected from objects such as clouds and other buildings, according to an implementation. These obstacles and reflections are not considered in the clear sky radiation calculations. Radiant light from the sky is determined based on sensor data from sensors such as infrared sensor 1534 , light sensor 1510 , and ambient temperature sensor 1536 of infrared cloud detector system 1502 . The tone level determined by module C is based on sensor data. In many cases, the hue rating is based on cloud cover conditions determined using readings from sensors. In general, the operation of module B will determine a tone level that darkens (or does not change) the tone level determined by module A, and the operation of module C will determine a tone level that brightens (or does not change) the tone level determined by module B.

The control logic may implement one or more of logic modules A, B, and C for each electrochromic window 1505 in the building or for a representative window division of a zone of electrochromic windows. Each electrochromic window 1505 may have a unique set of dimensions, orientations (eg, vertical, horizontal, tilted at an angle), positions, associated spatial types, and the like. A configuration file with this and other information can be maintained for each electrochromic window 1505 . The configuration file may be stored in a computer readable medium of the local window controller 1550 of the electrochromic window 1505 or in a building management system (BMS). The configuration file may contain information such as window configuration, occupancy lookup tables, information about associated reference glazing, and/or other data used by the control logic. The window configuration may include information such as the size of the electrochromic window 1505 , the orientation of the electrochromic window 1505 , the position of the electrochromic window 1505 , and the like. The occupancy lookup table describes the shade levels that provide occupant comfort for certain space types and depths of penetration. That is, the tint levels in the occupancy lookup table are designed to provide occupants in room 1500 with comfort from direct sunlight on the occupants or their workspace. Space type is a measure that determines how much tinting will be needed to address occupant comfort concerns and/or provide comfortable natural lighting in a room for a given depth of penetration. The Space Type parameter can take many factors into account. Among these factors are the type of work or other activity being performed in a particular room and the location of the activity. Close Work associated with detailed studies that require a lot of attention may be in one space type, while a break room or meeting room may have a different space type. Additionally, the location of a desk or other work surface in a room relative to a window is a consideration in defining a space type. For example, a space type may be associated with a single occupant office with a desk or other workspace located near the electrochromic window 1505 . As another example, the space type may be a foyer.

In some embodiments, one or more modules of the control logic may determine a desired tint level while considering energy savings in addition to occupant comfort. These modules can determine the energy savings associated with a particular tint level by comparing the performance of the electrochromic window 1505 at that tint level to a baseline glass or other standard reference window. The purpose of using this reference window may be to ensure that the control logic complies with the requirements of city building regulations or other requirements for reference windows used in building sites. Municipalities often use conventional low-emissivity glass to define reference windows to control the amount of air conditioning load in buildings. As an example of how the reference window 1505 fits into the control logic, the logic can be designed so that the irradiance coming through a given electrochromic window 1505 is never greater than the maximum irradiance coming through a reference window as specified by the respective municipality. In disclosed embodiments, the control logic may use the solar heat gain coefficient (SHGC) value of the electrochromic window 1505 at a particular tint level and the SHGC of the reference window to determine energy savings using that tint level. Typically, the value of SHGC is the fraction of incident light of all wavelengths transmitted through the window. While a reference glass is described in many embodiments, other standard reference windows may be used. In general, the SHGC of a reference window (eg, reference glass) is a variable that can vary for different geographic locations and window orientations, and is based on ordinance requirements specified by individual municipalities.

Typically, buildings are designed with a heating, ventilation and air conditioning ("HVAC") system with the capacity to meet the maximum expected heating and/or air conditioning load required in any given situation. The calculation of required capacity may take into account the datum glazing or reference windows required in the building at the particular location where the building is being constructed. Therefore, it is important that the control logic meet or exceed the functional requirements of the baseline glass in order to allow building designers to confidently determine how much HVAC capacity to place in a particular building. Since the control logic can be used to tint the windows to provide additional energy savings over baseline glass, the control logic can be useful to allow building designers to have lower HVAC capacity than would have been required using baseline glass specified by codes and standards.

Certain embodiments described herein assume that energy savings are achieved by reducing air conditioning loads in buildings. Therefore, many implementations attempt to achieve the maximum possible shading, taking into account occupant comfort levels and possible lighting loads in the room with the window in question. However, in some climates, such as those at more northern and southern latitudes, heating can be a greater concern than air conditioning. Therefore, the control logic can be modified, in particular in some cases reversed, so that less coloration occurs in order to ensure a reduction in the heating load of the building.

- Examples of control logic including modules A, B and C

16 depicts a flowchart 1600 showing the general control logic of a method for controlling one or more electrochromic windows in a building (eg, electrochromic window 1505 in FIGS. 15A - 15C ), according to an embodiment. The control logic calculates a tint level for the window using one or more of modules A, B, and C and sends instructions to transition the electrochromic window to that tint level. In operation 1610 , calculations in the control logic are performed 1 to n times at intervals clocked by a timer. For example, the hue level may be recalculated 1 to n times by one or more of modules A, B, and C, and calculated for time t _i = t ₁ , t ₂ . . . t _n . n is the number of recalculations performed, and n may be at least one. In some cases, logical calculations may be performed at constant time intervals. In one instance, logic calculations may be performed every 2 to 5 minutes. However, tint transitions for large pieces of electrochromic glass (eg, up to 6 feet by 10 feet) can take 30 minutes or more. For such large windows, calculations may be made on a less frequent basis, such as every 30 minutes.

In operation 1620 , logic modules A, B, and C perform calculations to determine a tint level for each electrochromic window at a single time t _i . These calculations may be performed by the processor of the controller. In some embodiments, the control logic is predictive logic that calculates how much the window should transition before actually transitioning. In such cases, the calculations in modules A, B, and C are _based on future times (e.g., ti = present time + duration of transition time such as electrochromic windows), e.g., approximately when the transition is complete or after the transition is complete. For example, the future time used in the calculation may be sufficiently in the future to allow the transition to be completed after the hue command is received. In such cases, the controller can send hue commands at the current time before the actual transition. When the transition is complete, the window will have transitioned to a desired tint level for that future time.

At operation 1630 , the control logic allows for certain types of alterations that depart from the algorithms at modules A, B, and C and at operation 1640 defines the altered tint level based on some other consideration. One type of modifier is a manual modifier. This is a modification implemented by the end user who has self-occupied the room and has determined that a particular tint level (modification value) is desirable. There may be a situation where the user's manual modification is modified by itself. An example of a modification is a high demand (or peak load) modification that is associated with a practical requirement for reduced energy consumption in a building. For example, on certain hot days in large urban areas, it may be necessary to reduce energy consumption throughout the urban area so as not to overburden the energy generation and delivery systems in the urban area. In such cases, the building can alter the tint level from the control logic described herein to ensure that all windows have a particularly high tint level. Another example of a modification might be if there is no occupant in the room, eg, in a commercial office building on weekends. In such cases, the building can be disengaged from one or more modules pertaining to occupant comfort, and all windows can have a low tinting rating in cold weather and a high tinting rating in warm weather.

At operation 1650 , control signals for implementing tint grading are transmitted over a network to a power supply in electrical communication with the electrochromic devices in one or more electrochromic windows in the building. In certain embodiments, transmission of tint levels to all windows of a building may be implemented with efficiency in mind. For example, if the recalculation of the tone level suggests that no change in tone from the current tone level is required, then there is no transmission of an instruction with an updated tone level. As another example, a building may be divided into zones based on window size and/or location in the building. In one case, the control logic recalculates the tone level more frequently for zones with smaller windows than for zones with larger windows.

In some embodiments, the control logic in FIG. 16 used to implement the control method for multiple electrochromic windows throughout a building can be on a single device (eg, a single master window controller). This device can perform calculations for each tintable window in a building, and also provides an interface for communicating tint levels to one or more electrochromic devices in an individual electrochromic window, for example, in a multi-zone window or across multiple EC lenses of an insulating glass unit. Some examples of multi-zone windows can be found in International PCT Application No. PCT/US14/71314, filed December 14, 2014, and entitled "MULTI-ZONE EC WINDOWS," which is hereby incorporated by reference in its entirety.

Also, there may be some adaptive components of the control logic of certain embodiments. For example, the control logic can determine how end users, such as occupants, attempt to alter the algorithm at certain times of day and use this information in a more predictive manner to determine desired tint levels. In one instance, the end user can use a wall switch to alter the tint level provided by the control logic at a certain time of day to an altered value. The control logic can receive information about these instances and alter the control logic to change the tint level to a modified value at that time of day.

FIG. 17 is a diagram showing a particular implementation of block 1620 from FIG. 16 . This figure shows the method of executing all three modules A, B and C in sequence to calculate the final tint level for a particular electrochromic window at a single time t _i . In the case of predictive logic, modules A, B, and C are executed based on determining the final tone level at a time in the future. The final tint level may be the maximum allowable transmittance of the window under consideration. Figure 17 also shows some exemplary inputs and outputs of modules A, B and C. Computations in modules A, B, and C are performed by processors in the local window controller, network controller, or host controller. While certain examples describe using all three modules A, B, and C, other implementations may use one or more of modules A, B, and C or may use additional/different modules.

At operation 1770 , the processor uses module A to determine a tint level for occupant comfort to prevent direct glare from sunlight from penetrating into the room. The processor uses module A to calculate the penetration depth of direct sunlight into the room based on the position of the sun in the sky and the window configuration from the configuration file. The position of the sun is calculated based on the latitude and longitude of the building and the time and date of the day. The occupancy lookup table and space type are inputs from a configuration file for that particular window. Module A outputs tone levels from A to module B. The goal of Module A is usually to ensure that direct sunlight or glare does not illuminate the occupants or their workspace. Determining the hue level from module A accomplishes this. Subsequent calculations of hue levels in modules B and C may reduce energy consumption and may require even larger hues. However, if a subsequent calculation of the tint level based on energy consumption suggests less tint than is needed to avoid disturbing the occupants, the logic prevents execution of the calculated larger level of transmittance to ensure occupant comfort.

At operation 1780 , the tone levels calculated in module A are input into module B. Typically, module B determines a tone level that darkens (or does not change) the tone level calculated in module B. The hue scale is calculated based on a calculation of irradiance under clear sky conditions (clear sky irradiance). The processor of the controller uses module B to calculate the clear sky irradiance for the electrochromic window based on the window orientation from the configuration file and based on the latitude and longitude of the building. These calculations are also based on the time and date of the day. Publicly available software such as the RADIANCE program (which is an open source program) may provide calculations for calculating clear sky irradiance. The SHGC of the reference glass is also imported into module B from the configuration file. The processor uses module B to determine a tint class that is darker than the tint class in A and transmits less heat than the reference glass calculated to transmit at maximum clear-air irradiance. The maximum clear sky irradiance is the highest irradiance level for all time calculated for clear sky conditions.

At operation 1790 , the tint level and calculated clear sky irradiance from module B are input into module C. The sensor readings are input to module C based on the measurement results obtained by the infrared sensor, the ambient temperature sensor and the light sensor. The processor uses module C to determine cloud cover conditions based on sensor readings and actual irradiance. The processor also uses module C to calculate the irradiance transmitted into the room if the window is tinted to the tint level from module B under clear sky conditions. If the actual irradiance through the window with the appropriate tint level is less than or equal to the irradiance through the window with the tint level from module B based on the determined cloud cover conditions from the sensor readings, then the processor uses module C to find the appropriate tint level. Typically, the operation of module C will determine a tone level that brightens (or does not change) the tone level determined by the operation of module B. In this example, the tone level determined in module C is the final tone level.

Much of the information input to the control logic is determined from fixed information about latitude and longitude, time of day, and date. This information describes the condition of the sun relative to the building, and more specifically relative to the windows on which the control logic is being implemented. The position of the sun relative to the window provides information such as the penetration depth of direct sunlight into the room through the window. It also provides an indication of the maximum irradiance or solar energy flux arriving through the window. This calculated irradiance level may be based on sensor inputs that may indicate that there is a decrease based on a determined cloud cover condition or another obstruction between the window and the sun.

A program such as the open source program Radiance is used to determine the clear sky irradiance based on the window orientation and the latitude and longitude coordinates of the building for a single _instant ti and the maximum value over all times. The reference glass SHGC and the calculated maximum clear sky irradiance are input into module B. Module B increases the hue grade calculated in module A step by step, and selects a hue grade whose internal radiation is less than or equal to the reference internal irradiance, where: internal irradiance = hue grade SHGC × clear sky irradiance, and reference internal irradiance = reference SHGC × maximum clear sky irradiance. However, when module A calculates the maximum tint of the glass, module B does not change the tint to make it lighter. The tone levels calculated in module B are then input into module C. The calculated clear sky irradiance is also input into module C.

- Example of control logic for making shading decisions using an infrared cloud detector system with a light sensor

FIG. 18 is a flowchart 1800 depicting one particular implementation of the control logic for the operations shown in FIG. 16 , according to an implementation. Although this control logic is described in relation to a single window, it should be understood that the control logic may be used to control multiple windows or a zone of one or more windows.

在操作1810 ，所述控制邏輯判定當日時間是否在以下時間週期中之一者期間：(i)開始於日出前不久之第一時間(例如，開始於日出前45分鐘、日出前30分鐘、日出前20或日出前之其他合適時間量之第一時間)與結束於日出後不久之第二時間(例如，結束於日出後45分鐘、日出後30分鐘、日出後20分鐘或日出後之其他合適時間量之第二時間)的時間週期；(ii)開始於日落前之一第三時間與結束於日落的時間週期；(iii)開始於日出後之第二時間後與結束於日落前之第三時間或結束於日落(黃昏)(例如，結束於日落前45分鐘、日落前30分鐘、日落前20分鐘、日落前0分鐘(亦即，在日落時)或日落前之其他合適時間量之第三時間)的時間週期；及(iv)開始於第三時間與結束於日出前之第一時間的時間週期。 In one instance, the time of sunrise can be determined from measurements taken by the visible wavelength light sensor. For example, the second time may be at the point when a visible light wavelength photosensor starts to measure direct sunlight, ie, the point when the intensity reading of the visible light photosensor is at or above a minimum intensity value. Additionally or alternatively, the third time may be determined to end at the point at which the intensity reading from the visible wavelength light sensor is at or below a minimum intensity value. In another example, the time of sunrise and/or the time of sunset can be calculated using a solar calculator and the day of the year, and time periods (i) to (iv) can be calculated by a defined period of time (e.g., 45 minutes) around the calculated time of sunrise/sunset.

If it is determined at operation 1810 that the time of day is not during one of time periods (i), (ii) or (iii), then the control logic determines that the time of day is in time period (iv) after time period (iii) and before time period (i). In this case, the control logic communicates the nighttime tint status (e.g., "clear" tint status or dark tint status, for safety), and proceeds to operation 1870 to determine whether there is a change, e.g., a change command received in the signal from an operator or occupant. If it is determined in operation 1860 that there is a change, the change value is the final tone level. If it is determined that there is no change in place, the nighttime tint state is the final tint level. At operation 1870 , a control command is sent via a network or directly to the window's electrochromic device to transition the window to the final tint level, the time of day is updated, and the method returns to operation 1810 .

If instead at operation 1810 it is determined that the time of day is during one of time periods (i), (ii) or (iii), then the time of day is between immediately before sunrise and immediately before or at sunset, and the control logic continues at operation 1820 to determine whether the solar azimuth is between the critical angles for tintable windows. If the control logic determines at operation 1820 that the sun azimuth is outside the critical angle, module A is bypassed and a "sunny" tint level is passed to module B, which is used for calculation at operation 1840 . If it is determined at operation 1820 that the sun azimuth is between the critical angles, then the control logic in module A is used at operation 1830 to calculate the penetration depth and calculate an appropriate hue level based on the penetration depth. The hue level determined from module A is then input to module B, and module B is used for the calculation at operation 1840 .

At operation 1840 , the control logic from module B determines a tone level that darkens (or does not change) the tone level it received from module A or the "sunny" tone level from operation 1820 . The hue scale is calculated in module B based on the calculation of irradiance under clear sky conditions (clear sky irradiance). Module B is used to calculate the clear sky irradiance of the window based on the window orientation from the configuration file and based on the latitude and longitude of the building. These calculations are also based on the time of day and day of the year. Publicly available software such as the RADIANCE program (which is an open source program) may provide calculations for determining clear sky irradiance. The SHGC of the reference glass is also imported into module B from the configuration file. The processor uses the control logic of module B to determine a tint level darker than (or the same as) the tint level it receives and transmits less heat than the reference glass calculated to transmit at maximum clear sky irradiance. The maximum clear sky irradiance is the highest irradiance level for all time calculated for clear sky conditions.

At operation 1850 , the tint level from module B, the calculated clear sky irradiance and sensor readings from the infrared sensor, ambient temperature sensor and light sensor are input to module C. The control logic of module C determines the cloud cover condition based on the sensor readings, and determines the actual irradiance based on the cloud cover condition. The control logic of module C also calculates an irradiance level that would be transmitted into the room if the window was tinted to the tint level from module B under clear sky conditions. If the actual irradiance determined through the window based on cloud cover conditions is less than or equal to the calculated irradiance through the window when tinted to the tint level from module B, the control logic in module C lowers the tint level. Typically, the operation of module C will determine a tone level that brightens (or does not change) the tone level determined by the operation of module B.

At operation 1850 , the control logic determines a tint level from module C based on the sensor readings, and then proceeds to operation 1860 to determine whether there is a modification in place, eg, a modification command received in a signal from the operator. If it is determined in operation 1860 that there is a change, the change value is the final tone level. If it is determined that there are no alterations in place, then the hue grade from module C is the final hue grade. At operation 1870 , a control command is sent or directed to the window's electrochromic device via a network to transition the window to the final tint level, the time of day is updated, and the method returns to operation 1810 .

FIG. 19 is a flow diagram 1900 depicting one particular implementation of the control logic implementing operations 1850 of module C as shown in FIG. 18 . At operation 1910 , one or more signals are received at the processor having a temperature reading T _{IR taken by an infrared sensor at a particular sample time,} a temperature reading T _amb taken by an ambient temperature sensor at a sample time, and an intensity reading taken by a light sensor at a sample time. Signals from the infrared sensor, the ambient temperature sensor and the light sensor are received wirelessly and/or via a wired electrical connection. The infrared sensor takes temperature readings based on the infrared radiation it receives within its field of view. Infrared sensors are typically oriented toward an area of the sky of interest (eg, an area above a building with windows). The ambient temperature sensor is configured to be exposed to the environment outside the building to measure the ambient temperature. Ambient temperature sensors are typically positioned and their sensing surfaces are oriented such that direct sunlight is blocked or diffused without illuminating the sensing surface. Direct sunlight is typically diffused (eg, by a diffuser) before striking the sensing surface of the photosensor. In some cases, the sensing surface of the light sensor is oriented in the same direction as the window facing. If it is determined in operation 1920 that the time of day is during any of the time periods (i) or (iii), the processor calculates the difference (Δ) between the temperature reading T _IR obtained by the infrared sensor and the temperature reading T _amb obtained by the ambient temperature sensor at a sample time (operation 1930 ). Optionally (represented by the dotted line), a correction factor is applied to the calculated delta (Δ) (operation 1930 ). Some examples of applicable correction factors include humidity, sun angle/elevation, and site elevation.

In one embodiment, the processor also determines at operation 1920 whether the infrared readings oscillate at a frequency greater than a second defined level. If the processor determines at operation 1920 that the time of day is within time period (i) or (iii) and the infrared readings oscillate at a frequency greater than a second defined level, the processor applies operation 1990 to determine cloud conditions using the light sensor readings. For example, the processor may determine a "clear" condition if the light sensor reading is above a certain minimum intensity level, and a "cloudy" condition if the light sensor reading is at or below the minimum intensity level. Control logic then applies operation 1995 to determine a hue level based on the determined cloud condition. If the infrared cloud detector is still being implemented and in operation, the method advances to the next sample time and returns to operation 1910 . Otherwise, the method returns to operation 1860 in FIG. 18 .

In operation 1934 , the processor determines whether the calculated delta (Δ) value is below a threshold value (eg, -5 millidegrees Celsius, -2 millidegrees Celsius, etc.). If it is determined that the calculated delta (Δ) value is lower than the lower threshold value, then the cloud cover condition is determined to be a "clear" condition (operation 1936 ). Control logic then applies operation 1995 to determine a hue level based on the determined cloud condition. During operation/implementation of the infrared cloud detector, the method then increments to the next sample time and returns to operation 1910 . If an infrared cloud detector is not being implemented, the method returns to operation 1860 in FIG. 18 .

If it is determined that the calculated difference (Δ) is higher than the lower threshold, the processor determines in operation 1940 whether the calculated difference (Δ) is higher than an upper threshold (eg, 0 millidegrees Celsius, 2 millidegrees Celsius, etc.). If it is determined at operation 1940 that the calculated delta (Δ) is above the upper threshold, then the processor determines that the cloud coverage condition is a "cloudy" condition (operation 1942 ) and applies operation 1995 to determine the hue level based on the determined cloud condition. During operation/implementation of the infrared cloud detector, the method then increments to the next sample time and returns to operation 1910 . If an infrared cloud detector is not being implemented, the method returns to operation 1860 in FIG. 18 .

At operation 1995 , the control logic determines the actual irradiance based on cloud cover conditions and calculates an irradiance level that would be transmitted into the room if the window were tinted to the tint level from module B under clear sky conditions. Control logic in module C typically lowers the tint level from module B if the irradiance based on cloud cover conditions is less than or equal to the calculated irradiance through the window when tinted to the tint level from module B. During operation/implementation of the infrared cloud detector, the method then increments to the next sample time and returns to operation 1910 . If the infrared cloud detector is still not being implemented, the method returns to operation 1860 in FIG. 18 .

If it is determined at operation 1940 that the calculated delta (Δ) is below the upper threshold, then the processor determines the cloud coverage condition as "intermittently cloudy" or another intermediate condition (operation 1950 ) and proceeds to operation 1995 , as described in detail above.

If it is determined at operation 1920 that the time of day is not during either time period (i) or (iii), then the time of day is during time period (ii) (i.e., "daylight"), and at operation 1970 , the processor calculates the difference between the temperature reading T _IR taken by the infrared sensor and the intensity reading taken by the light sensor. At operation 1980 , the processor determines that the calculated difference is within an acceptable limit. If the processor determines at operation 1980 that the calculated difference is greater than the acceptable limit, then the processor applies operation 1930 to calculate the difference (Δ) and use the calculated difference (Δ) to determine the cloud cover condition, as discussed above.

In one embodiment, the processor also determines at operation 1960 whether the infrared readings oscillate at a frequency greater than a second defined level. If the processor determines at operation 1960 that the time of day is within time period (ii) and the infrared readings oscillate at a frequency greater than a second defined level, the processor applies operation 1990 to determine cloud conditions using the light sensor readings. For example, the processor may determine a "clear" condition if the light sensor reading is above a certain minimum intensity level, and a "cloudy" condition if the light sensor reading is at or below the minimum intensity level. The control logic then proceeds to operation 1995 described in detail above.

If the processor determines at operation 1980 that the calculated difference is within acceptable limits, then the light sensor readings are used to determine cloud cover conditions (operation 1990 ). For example, the processor may determine a "clear" condition if the light sensor reading is above a certain minimum intensity level, and a "cloudy" condition if the light sensor reading is at or below the minimum intensity level. The control logic then proceeds to operation 1995 described in detail above.

In one embodiment, the processor also determines at operation 1910 whether the light sensor readings oscillate at a frequency greater than a first defined level and the infrared readings oscillate at a frequency greater than a second defined level. If the processor determines at operation 1970 that the calculated difference is within acceptable limits and the processor determines that the light sensor reading oscillates at a frequency greater than the first defined level, then the processor applies operation 1930 to calculate the difference (Δ) and use the calculated difference (Δ) to determine cloud cover conditions, as discussed above. If the processor determines at operation 1970 that the calculated difference is not within acceptable limits and the processor determines that the infrared readings oscillate at a frequency greater than the second defined level, the processor applies operation 1990 to determine cloud conditions using the light sensor readings. For example, the processor may determine a "clear" condition if the light sensor reading is above a certain minimum intensity level, and a "cloudy" condition if the light sensor reading is at or below the minimum intensity level. Control logic then proceeds to operation 1995 described in more detail above.

- Examples of modules C1 and D

One example of exemplary tint control logic includes four (4) logic modules A, B, C1 and D. Modules C1 and D use sensor readings taken by various sensors, such as temperature sensors and visible light sensors, to determine hue levels. In one embodiment, modules C1 and D also use ambient temperature readings from weather feeds. Typically, these sensors are part of an infrared cloud detector system , such as in the form of one of a multi-sensor device (eg, multi-sensor device 2030 shown in FIGS. 20A - 20D , multi-sensor device 401 shown in FIGS. 4A - 4C , or multi-sensor device 3201 shown in FIGS. 32A - 32C ).

Figures 20A - 20D contain schematic diagrams depicting certain general inputs to logic modules A, B, C1 and D. To illustrate general input, each figure depicts a schematic side view of a room 2000 of a building with a desk 2001 and an electrochromic window 2005 between the exterior and interior of the building. The figure also depicts a terminal window controller 2050 in communication with the electrochromic window 2005 to send control signals to control the voltage applied to the electrochromic device of the electrochromic window 2005 to control its transition to different shade levels. The figure also depicts an infrared cloud detector system in the form of a multi-sensor device 2030 located on the roof of a building with one or more tintable windows.

The multi-sensor device 2030 is illustrated in simplified form in FIGS. 20A - 20D . The components of multi-sensor device 2030 are similar to the components of multi-sensor device 3201 described in more detail with respect to FIGS. 32A - 32C . In the illustrated example shown in FIGS. 20A - 20D , the multi-sensor device 2030 includes a housing 2032 having an interior made of a light diffusing material. The multi-sensor device 2030 also includes at least two redundant IR sensor devices 2034 , eg, multiple IR sensor devices, to provide redundancy if one IR sensor device fails or is unavailable. Each of the infrared sensor devices 2034 has an on-board ambient temperature sensor and an infrared sensor for measuring thermal radiation from the sky. In addition, the multi-sensor device 2030 includes a plurality of visible light sensors 2010 located within the casing of the housing and facing outward and/or upward in different directions. For example, the multi-sensor device 2030 can have thirteen (13) visible light sensors 2010 . Each infrared sensor of multi-sensor device 2030 is configured to take a temperature reading of the sky, T _sky , based on infrared radiation received from an area of the sky within its field of view. Each onboard ambient temperature sensor is configured to take an ambient temperature reading T _amb of the ambient air. Each infrared sensor device 2034 includes an imaginary axis (not shown) that is perpendicular to the sensing surface of the infrared sensor and passes approximately through the center of the sensing surface. Although not shown, multi-sensor device 2030 also includes one or more structures that retain its components within housing 2032 . Although logic modules A, B, C1, and D are described with reference to sensor data from multi-sensor device 2030 , for simplicity it should be understood that these modules may use data derived from one or more other sources, such as other infrared cloud detector systems, weather feeds, other sensors in the building (such as freestanding sensors available at one or more electrochromic windows), user input, etc.

Multi-sensor device 2030 also includes a processor that executes instructions stored in memory (not shown) for implementing logic. For example, in one implementation, the processor of the multi-sensor device 2030 filters the sensor readings using the logic of module D (eg, the logic of module D' described with reference to FIG. 23 ). In this example, the processor of the multi-sensor device 2030 receives sensor readings from sensors at the multi-sensor device 2030 and/or receives weather feeds over a network to filter the sensor readings over time to determine filtered sensor values as input to the control logic. In this implementation, the window controller 2050 receives the signal with the filtered sensor value and uses the filtered sensor value as an input into the logic of modules C1 and/or D.

Room 2000 also includes a local window controller 2050 having a processor (not shown) that can execute instructions stored in memory (not shown) for implementing control logic to control the tint levels of electrochromic windows 2005 . Window controller 2050 communicates with electrochromic window 2005 to send control signals. The window controller 2050 also communicates (wirelessly or wired) with the multi-sensor device 2030 to receive signals having, for example, filtered sensor values or sensor readings. For example, window controller 2050 may receive a signal having an infrared sensor reading ( T _sky ) obtained by an infrared sensor, an ambient temperature reading ( T _amb ) obtained by an on-board ambient temperature sensor of infrared sensor device 2034 and/or a visible light reading obtained by plurality of light sensors 2010 . Additionally or alternatively, window controller 2050 may receive signals having filtered infrared sensor values based on readings taken by infrared sensor 2034 and/or filtered light sensor values based on readings taken by light sensor 2010 .

According to certain aspects, power/communication lines run from a building or another structure to multi-sensor device 2030 . In one implementation, the multi-sensor device 2030 includes a network interface that can couple the multi-sensor device 2030 to a suitable cable. Multi-sensor device 2030 may communicate data via a network interface to window controller 2050 or another controller of the building (eg, a network controller and/or a master controller). In some other implementations, multi-sensor device 2030 may additionally or alternatively include a wireless network interface to enable wireless communication with one or more external controllers. In some aspects, multi-sensor device 2030 may also include a battery within or coupled to its housing to power the sensors and electrical components within it. Instead or in addition to power from a power supply (eg, from a building power supply), the battery can provide this power. In one aspect, multi-sensor device 2030 further includes at least one photovoltaic cell, eg, on an outer surface of the housing. The at least one photovoltaic cell may provide power instead or in addition to power provided by any other power supply.

FIG. 20A shows the penetration depth of direct sunlight into the room 2000 through the electrochromic window 2005 between the exterior and interior of the building containing the room 2000 . Penetration depth is measured as one of how far direct sunlight can penetrate into a room in 2000 . As shown, penetration depth is measured in the horizontal direction away from the sill (bottom) of the electrochromic window 2005 . Typically, the window defines an aperture that provides an acceptance angle for direct sunlight. Penetration depth is calculated based on the geometry of the window (e.g., window size), its location and orientation in the room, any fins or other external shading outside the window, and the position of the sun (e.g., the angle of direct sunlight at a particular time of day and on a particular day). External shading to the electrochromic window 2005 can be due to any type of structure that can shade the window, such as overhangs, fins, etc. In FIG. 20A , there is an overhang 2020 above the electrochromic window 2005 , which blocks a portion of the direct sunlight from entering the room 2000 , thus shortening the penetration depth.

Module A1 may be used to determine a tint level that takes into account occupant comfort from direct sunlight passing through the electrochromic window 2005 onto the occupant or their active area (also referred to herein as "glare conditions"). The tint level is determined based on the calculated penetration depth of direct sunlight into the room and the type of space in the room at a particular moment (time of day and day of year) (eg, table near a window, hallway, etc.). In some cases, the tint level determination may also be based on providing sufficient natural lighting into the room. In some cases, the depth of penetration is calculated at a future time to account for the vitrification time (the time required for the window to tint, eg, to 80%, 90%, or 100% of the desired shade level). The problem addressed in module A1 is that direct sunlight can penetrate so deep into room 2000 that it falls directly on occupants working at a desk or other work surface in the room. The sun's position can be calculated using publicly available programs and allows easy calculation of penetration depth.

20A - 20D also show desk 2001 in room 2000 , as an example of a single occupancy office space type with desks associated with activity areas (ie, desks) and locations of activity areas (ie, desk locations). Each space type is associated with a different tone level for occupant comfort. For example, if the activity is a key activity such as work in an office being done at a desk or computer, and the desk is located near a window, then the desired tint level may be higher than if the desk is further away from the window. As another example, if the activity is non-critical, such as in the foyer, the desired tone level may be lower than in the same space with a desk for an office.

20B shows direct sunlight and radiation entering a room 2000 through an electrochromic window 2005 under clear sky conditions, according to an implementation. Radiation can come from sunlight scattered by molecules and particles in the atmosphere. Module B determines a tint level based on the calculated irradiance under clear sky conditions flowing through the electrochromic window 2005 under consideration. Various software such as the open source RADIANCE program can be used to calculate clear sky irradiance at a latitude, longitude, time of year, and time of day, and for a given window orientation.

20C shows radiant light from the sky as may be blocked by or reflected from objects such as clouds or other buildings or structures, according to an implementation. These obstacles and reflections are not considered in the clear sky radiation calculations. Radiant light from the sky is determined based on light sensor data from the plurality of light sensors 2010 of the multi-sensor device 2030 . The tone level determined by the logic of module C1 is based on the light sensor data. The hue rating is based on cloud cover conditions determined using readings taken by a plurality of light sensors 2010 . In some cases, the cloud cover condition is determined based on a filtered light sensor value determined from readings from the plurality of light sensors 2010 taken over time.

FIG. 20D shows that infrared radiation 2090 from the sky can radiate from clouds and other obstructions, according to an implementation. As mentioned above with reference to Figure 20C , such obstacles are not considered in the clear sky radiation calculations. Operation of module D uses sky temperature and ambient temperature data to determine cloud cover conditions during morning and evening periods when visible light levels are low and visible light sensor readings are low and may give false positives for cloudy conditions.

In one implementation, the operation of module D determines cloud cover conditions at each time t _i using filtered infrared sensor values determined based on sky temperature readings ( T _sky ) and ambient temperature readings ( T _amb ) taken over time. Ambient temperature readings come from one or more ambient temperature sensors or from weather feeds. For example, the sky temperature reading may be determined based on readings taken by the infrared sensors of the multi-sensor device 2030 . Hue ratings are based on cloud cover conditions determined from filtered infrared sensor values. Typically, the operation of module B will determine a tone level that darkens (or does not change) the tone level determined by module A1, and the operation of module C1 or D will determine a tone level that brightens (or does not change) the tone level determined by module B.

The control logic may implement one or more of logic modules A1, B, C1 and D for one or more windows in the building. Each electrochromic window may have a unique set of dimensions, orientation (eg, vertical, horizontal, tilted at an angle), position, associated spatial type, and the like. A configuration file with this and other information may be maintained for each electrochromic window or zone of electrochromic windows in the building. In one example, the configuration file may be stored in a computer readable medium of the local window controller 2050 of the electrochromic window 2005 or in a building management system (BMS). The configuration file may contain information such as window configuration, occupancy lookup tables, information about associated reference glazing, and/or other data used by the control logic. The window configuration may include information such as the size of the electrochromic window 2005 , the orientation of the electrochromic window 2005 , the position of the electrochromic window 2005 , and the like. The occupancy lookup table describes the shade levels that provide occupant comfort for certain space types and depths of penetration. That is, the tint levels in the occupancy lookup table are designed to provide occupants in room 2000 with comfort from direct sunlight on the occupants or their workspace. Space type is a measure that determines how much tinting will be needed to address occupant comfort concerns and/or provide comfortable natural lighting in a room for a given depth of penetration. The Space Type parameter can take into account many factors. Among these factors are the type of work or other activity being performed in a particular room and the location of the activity. Close Work associated with detailed studies that require a lot of attention may be in one space type, while a break room or meeting room may have a different space type. Additionally, the location of a desk or other work surface in a room relative to a window is a consideration in defining a space type. For example, a space type may be associated with a single occupant office with a desk or other workspace located near the electrochromic window 2005 . As another example, the space type may be a foyer.

In some embodiments, one or more modules of the control logic may determine a desired tint level while considering energy savings in addition to occupant comfort. These modules can determine the energy savings associated with a particular tint level by comparing the performance of the electrochromic window at that tint level to a baseline glass or other standard reference window. The purpose of using this reference window may be to ensure that the control logic complies with the requirements of city building regulations or other requirements for reference windows used in building sites. Municipalities often use conventional low-emissivity glass to define reference windows to control the amount of air conditioning load in buildings. As an example of how the reference window 2005 fits into the control logic, the logic can be designed so that the irradiance coming through a given electrochromic window 2005 is never greater than the maximum irradiance coming through a reference window as specified by the respective municipality. In disclosed embodiments, the control logic may use the solar heat gain coefficient (SHGC) value of the electrochromic window 2005 at a particular tint level and the SHGC of the reference window to determine energy savings using that tint level. Typically, the value of SHGC is the fraction of incident light of all wavelengths transmitted through the window. While a reference glass is described in many embodiments, other standard reference windows may be used. In general, the SHGC of a reference window (eg, reference glass) is a variable that can vary for different geographic locations and window orientations, and is based on ordinance requirements specified by individual municipalities.

- Examples of control logic including modules A, B, C1 and D

21 depicts a flowchart 2100 showing the general control logic of a method for controlling one or more electrochromic windows in a building, according to an embodiment. For example, such control logic can be implemented to control the zones of one or more electrochromic windows. The control logic implements one or more of modules A1, B, C1, and D to calculate a tint level for the one or more electrochromic windows, and to send instructions to transition the electrochromic device of the one or more electrochromic windows (e.g., an electrochromic device in a multi-zone electrochromic window or an electrochromic device on multiple electrochromic lenses of an insulating glass unit) to the tint level. Some examples of multi-zone windows can be found in International PCT Application No. PCT/US14/71314, filed December 14, 2014, and entitled "MULTI-ZONE EC WINDOWS," which is hereby incorporated by reference in its entirety. Modules A1 and B are similar to modules A and B described with respect to Figures 15A and 15B.

Calculations in the control logic are performed at intervals ticked by the timer at operation 2110 . In some cases, logical calculations may be performed at constant time intervals. In one case, logical calculations are performed every 2 to 5 minutes. In other cases, calculations may need to be made on a less frequent basis, such as every 30 minutes or every 20 minutes, such as for tint transitions for large electrochromic lenses (e.g., up to 6 feet by 10 feet) that can take 30 minutes or more to change.

At operation 2120 , logic modules A1, B, C1, and D perform calculations to determine the tint level for one or more electrochromic windows at a single time t _i . These calculations may be performed by one or more processors of the window controller and/or the multi-sensor device. For example, a processor of a multi-sensor device may determine filtered sensor values and communicate these filtered sensor values to a window controller that determines tint levels based on the filtered sensor values. In another example, one or more processors of the window controller may determine filtered sensor values and corresponding tint levels based on sensor readings received from a multi-sensor device.

In some embodiments, the control logic is predictive logic that calculates how much the window should transition before actually transitioning. In such cases, the calculations in modules A1, B, C1, and D are based on a future time (e.g., ti = present time + duration such as the transition time of one or more electrochromic windows), e.g., approximately when the transition _is complete or after the transition is complete. For example, the future time used in the calculation may be sufficiently in the future to allow the transition to be completed after the hue command is received. In such cases, the window controller can send tint commands at the current time before the actual transition. When the transition is complete, the window will have transitioned to a desired tint level for that future time.

At operation 2130 , the control logic allows various types of alterations to depart from the algorithms at modules A1, B, C1, and D, and at operation 2140 defines the altered tint levels based on some other consideration. One type of modifier is a manual modifier. This is a modification implemented by the end user who has self-occupied the room and has determined that a particular tint level (modification value) is desirable. There may be a situation where the user's manual modification is modified by itself. An example of a modification is a high demand (or peak load) modification that is associated with a practical requirement for reduced energy consumption in a building. For example, on certain hot days in large urban areas, it may be necessary to reduce energy consumption throughout the urban area so as not to overburden the energy generation and delivery systems in the urban area. In such cases, the building can alter the tint level from the control logic to ensure that all windows have a particularly high tint level. This modifier overrides the user's manual modifier. Another example of alteration is when there are no occupants in the room, eg, in a commercial office building on weekends. In this case, the building can be disengaged from one or more modules pertaining to occupant comfort, and all windows can have a low tinting rating in cold weather and a high tinting rating in warm weather.

At operation 2150 , control signals for implementing the tint level are transmitted over a network to a power supply in electrical communication with the electrochromic devices of the one or more electrochromic windows to transition the window to the tint level. In certain embodiments, the transmission of the tint levels to the windows of buildings may be implemented with efficiency in mind. For example, if the recalculation of the tone level suggests that no change in tone from the current tone level is required, then there is no transmission of an instruction with an updated tone level. As another example, a building may be divided into zones of windows based on window size and/or location in the building. In one case, the control logic recalculates the tone level more frequently for zones with smaller windows than for zones with larger windows.

In one case, the control logic in FIG. 21 implements a control method for controlling the tint levels of all electrochromic windows for all buildings on a single device (eg, on a single master window controller). This device can perform calculations for each electrochromic window in the building, and also provides an interface for transmitting the tint level to the electrochromic device in the individual electrochromic window.

Also, there may be some adaptive components of the control logic of certain embodiments. For example, the control logic may determine how end users (eg, occupants) attempt to alter the algorithm at certain times of day and then use this information in a more predictive manner to determine a desired tint level. For example, an end user could use a wall switch to switch the number of The tone level provided at a certain time above is changed to a changed value. Control logic may receive information about these instances and alter the control logic to introduce a modification value from the end user at that time of day that changes the tint level to the modification value.

FIG. 22 is a diagram showing a particular implementation of block 2020 from FIG. 21 . This figure shows the method of executing all four modules A1, B, C1 and D sequentially to calculate the final tint level for a particular electrochromic window at a single time t _i . In the case of predictive logic, modules A1, B, C1 and D are executed based on determining the final tone level at a time t _i in the future. The final tint level may be the maximum allowable transmittance of the window under consideration. In one embodiment, the calculations of the modules A1, B, C1 and D are performed by the processor of the local window controller, network controller or main controller.

At operation 2270 , the processor uses module A1 to determine a tint level for occupant comfort to prevent direct glare from sunlight from penetrating into the room. The processor uses module A1 to calculate the penetration depth of direct sunlight into the room based on the position of the sun in the sky and the window configuration from the configuration file. The position of the sun is calculated based on the latitude and longitude of the building and the time and date of the day. The occupancy lookup table and space type are inputs from a configuration file for that particular window. Module A1 outputs the tone grades from A1 to module B. The goal of module A1 is generally to ensure that direct sunlight or glare does not illuminate occupants or their workspaces. Determining the tone level from module A1 accomplishes this. Subsequent calculations of hue levels in modules B, C1 and D may reduce energy consumption and may require even larger hues. However, if a subsequent calculation of the tint level based on energy consumption suggests less tint than is needed to avoid disturbing the occupants, the logic prevents execution of the calculated larger level of transmittance to ensure occupant comfort.

In operation 2280 , the tone level calculated in module A1 is input into module B. Typically, module B determines a tone level that darkens (or does not change) the tone level calculated in module B. The hue scale is calculated based on a calculation of irradiance under clear sky conditions (clear sky irradiance). The controller's processor uses module B to calculate the clear sky irradiance for one or more electrochromic windows based on the window orientation from the configuration file and based on the latitude and longitude coordinates of the building. These calculations are also based on the time _of day at time ti and/or the maximum value for all times. Publicly available software such as the RADIANCE program (which is an open source program) may provide calculations for calculating clear sky irradiance. The SHGC of the reference glass is also imported into module B from the configuration file. The processor uses module B to determine a tint class that is darker than the tint class in A1 and transmits less heat than the reference glass calculated to transmit at maximum clear-air irradiance. The maximum clear sky irradiance is the highest irradiance level for all time calculated for clear sky conditions. In one example, module B steps up the hue grade calculated in module A1 and selects a hue grade whose internal irradiance is less than or equal to the reference internal irradiance, where: internal irradiance = hue grade SHGC x clear sky irradiance, and reference internal irradiance = reference SHGC x maximum clear sky irradiance.

At operation 2290 , the hue level and light sensor readings and/or filtered light sensor values from module B are input to module C1. The calculated clear sky irradiance is also input into module C1. Light sensor readings are based on measurements taken by multiple light sensors, such as a multi-sensor device. The processor uses the logic of module C1 to determine a cloud cover condition by comparing filtered light sensor values to thresholds. In one instance, module C1 determines filtered light sensor values based on raw light sensor readings. In another case, module C1 receives filtered light sensor values as input. The processor implements the logic of module C1 to determine a hue level based on the determined cloud cover condition. Typically, the operation of module C1 will determine a tone level that brightens or does not change the tone level determined by the operation of module B.

At operation 2295 , the tone level from module C1 is input to module D. Additionally, the IR sensor reading and the ambient temperature sensor reading and/or their associated filtered IR sensor values are input to module D. The infrared sensor reading and the ambient temperature sensor reading comprise the sky temperature reading ( T _sky ), the ambient temperature reading from the local sensor at the building ( T _amb ) or the ambient temperature reading from the weather feed ( T _weather ) and/or the difference between T _sky -T _amb . Determine _the filtered IR sensor value based on the sky temperature reading ( T _sky ) and the ambient temperature reading from the local sensor ( Tamb ) or the ambient temperature reading from the weather feed ( T _weather ). Sky temperature readings are taken by one or more infrared sensors.

Ambient temperature readings may be received from various sources. For example, ambient temperature readings may be communicated from one or more ambient temperature sensors located on an infrared sensor board and/or from a stand-alone temperature sensor such as a multi-sensor device at a building. As another example, ambient temperature readings may be received from weather feeds. The logic of module D determines the cloud cover condition by comparing the filtered infrared sensor value to a threshold value. Typically, the operation of module D will determine a tone level that darkens (or does not change) the tone level determined by the operation of module C1. In this example, the tone level determined in module D is the final tone level.

Much of the information input to the control logic described with respect to Figure 22 is determined from fixed information about the latitude and longitude of the building and also from the time and date of day (day of the year). This information describes the position of the sun relative to the building, and more specifically, relative to each of the one or more windows for which the control logic is being implemented. The position of the sun relative to the windows can be used to calculate information such as the penetration depth of direct sunlight into the room through each window. It also provides an indication of the maximum irradiance or solar radiant energy flux arriving through the window during clear sky conditions.

In the morning and evening, sunlight levels are low and readings taken by external visible light sensors, such as in a multi-sensor arrangement, are low, which readings can be considered consistent with readings taken during cloudy conditions during the day. For this reason, external visible light sensor readings taken during morning and evening may falsely indicate a cloudy condition if considered in isolation. Additionally, any obstruction from buildings or mountains/mountains could also result in a false positive indication for cloudy conditions based on the visible light sensor readings taken alone. Furthermore, external visible light sensor readings taken before sunrise, if taken alone, can lead to a false positive for cloudy conditions. Where the control logic predictively predetermines the hue level at sunrise based on a single visible light sensor reading taken immediately before sunrise, a false positive for cloudy conditions may result in switching the electrochromic window to a clear state at sunrise, causing glare in the room.

In some implementations, the control logic described herein uses The filtered sensor values of the outside sensor and temperature readings from the ambient temperature sensor determine cloud conditions in the morning and evening and/or at times immediately before sunrise. The one or more infrared sensors typically operate independently of sunlight levels, allowing the tinting control logic to determine cloud conditions before sunrise, and as the sun sets, determine and maintain proper tint levels during the morning and evening. Additionally, filtered sensor values based on temperature readings from one or more infrared sensors can be used to determine a cloud condition even when the visible light sensor is covered or otherwise blocked.

In one implementation, the control logic described with respect to FIG. 22 implements module C1 and/or module D based on whether time ti _is in the am, day, or night range (as determined by sun altitude). An example of this implementation is described in detail with respect to FIG. 24 .

- Examples of Module D and Module D'

In some implementations, module D uses filtered infrared (IR) sensor values (eg, a moving average or median of sensor readings) to determine a tint level for one or more electrochromic windows in a building. The filtered IR sensor values can be calculated by logic and passed to module D, or module D can query a database to retrieve stored filtered IR sensor values. In one aspect, module D includes logic to calculate a filtered IR sensor value using the cloudy bias value and the sky temperature reading ( T _sky ) and the ambient temperature reading from the local sensor ( T _amb ) or from the weather feed ( T _weather ) and/or the difference between the sky temperature reading and the ambient temperature reading (Delta (Δ)). The cloudy deviation value is a temperature deviation corresponding to the threshold value that will be used by the logic in module D to determine the cloudy condition. The logic of module D may be performed by one or more processors executing the logic of module D, such as a local window controller, network controller, or host controller.

For example, an alternative implementation of the control logic shown in FIG. 22 further includes a module D' that receives the IR sensor readings from the sensors and the ambient temperature reading, calculates the filtered IR sensor value, and communicates the filtered IR sensor value to module D. Alternatively, the logic of module D' may be executed by one or more processors of the multi-sensor device. In one case, will come from the mod The calculated filtered IR sensor value of D' is saved to a database of IR sensor measurement results stored in memory. In this case, the one or more processors executing the calculations of module D retrieves filtered IR sensor values from the IR sensor measurement database as input.

FIG. 23 illustrates a flowchart 2300 depicting the logic of module D' according to certain implementations. The logic of module D' can be executed by one or more processors of the local window controller, network controller, main controller or multi-sensor device. At operation 2310 , the processor performing the operations of module D' receives as input the sensor readings of the current time. Sensor readings may be received via a communication network at the building, for example, from a rooftop multi-sensor installation. The received sensor readings include the sky temperature reading ( T _sky ) and _the ambient temperature reading ( T _amb ) from the local sensor at the building or from the weather feed ( T _weather ) and/or the reading of the difference between T _sky and Tamb (Δ). The ambient temperature reading ( T _amb ) from the local sensor at the building is the measurement taken by the ambient temperature sensor on board or separate from the IR sensor device. Ambient temperature sensor readings may alternatively come from weather feeds.

In one implementation, the logic of module D' receives and uses raw sensor readings of measurements taken by two or more IR sensor devices at the building (e.g., of a rooftop multi-sensor device), each IR sensor device having an onboard ambient temperature sensor for measuring ambient temperature ( T _amb ) and an onboard infrared sensor directed toward the sky for measuring sky temperature ( T _sky ) based on infrared radiation received within its field of view. Two or more IR sensor devices are typically used to provide redundancy. In one case, each infrared sensor device outputs a reading of the ambient temperature ( T _amb ) and a reading of the sky temperature ( T _sky ). In another case, each infrared sensor device outputs a reading of ambient temperature ( T _amb ), a reading of sky temperature ( T _sky ), and a reading of _the difference (difference Δ) between T _sky and Tamb . In one case, each infrared sensor device outputs _a reading of the difference (difference Δ) between T _sky and Tamb . According to one aspect, the logic of module D' uses the raw sensor readings of the measurements taken by the two IR sensor devices at the building. In another aspect, the logic of module D' uses raw sensor readings of measurements taken by 1 to 10 IR sensor devices at the building.

In another implementation, the logic of module D' receives and uses raw sky temperature ( T _sky ) readings taken by infrared sensors at the building and directed towards the sky to receive infrared radiation within its field of view and ambient temperature readings ( T _weather ) from weather feeds. Weather feed data is received from one or more weather services and/or other data sources via a communication network. Weather feed data typically includes data associated with weather conditions such as percent cloud cover, visibility data, wind speed data, temperature data, percent chance of precipitation, and/or humidity. Typically the weather feed is received by the window controller in a signal via a communication network. According to some aspects, a window controller may send a signal with a request for weather feeds to one or more weather services over a communication interface via a communication interface. The request typically includes at least the latitude and longitude of the location of the window being controlled. In response, the one or more weather services send a signal with the weather feed to the window controller via a communication interface via the communication network. The communication interface and network can be wired or wireless. In some cases, weather services may be accessed via weather websites. An example of a weather website can be found at www.forecast.io . Another example is the National Weather Service (www.weather.gov). Weather feeds can be based on the current time, or can be forecast for a future time. More details on the logic of using weather feeds can be found in International Application PCT/US16/41344, filed July 7, 2016, entitled CONTROL METHOD FOR TINTABLE WINDOWS, which is hereby incorporated by reference in its entirety.

Returning to FIG. 23 , at operation 2320 , a temperature value ( T _calc ) is calculated based on sky temperature readings from one or more infrared sensors, ambient temperature readings from one or more local ambient temperature sensors or from weather feeds, and cloudy bias values. The cloudy deviation value is a temperature deviation used to determine the first and second threshold values for determining cloud conditions in module D. In one implementation, the cloudy bias value is -17 millidegrees Celsius. In one example, the cloudy deviation value of -17 degrees Celsius corresponds to the first threshold value of 0 degrees Celsius. In one implementation, the cloudy bias value is in the range of -30 millidegrees Celsius to 0 millidegrees Celsius.

In one implementation, the temperature value ( T _calc ) is calculated based on sky temperature readings from two or more pairs of thermal sensors. Each pair of heat sensors has an infrared sensor and an ambient temperature sensor. In one instance, the thermal sensors of each pair are integral components of the IR sensor device. Each IR sensor device has an onboard infrared sensor and an onboard ambient temperature sensor. Two IR sensor devices are typically used to provide redundancy. In another instance, the infrared sensor is separate from the ambient temperature sensor.

In one implementation, the temperature value is calculated as: T _calc =minimum( T _sky1 ,T _{sky2 ,} ...)-minimum( T _amb1 ,T _amb2 ,...)-cloudy bias (Equation 2)

T _sky1 , T _sky2 . . . are temperature readings obtained by multiple infrared sensors, and T _amb1 , T _{amb2 .} . . are temperature readings obtained by multiple ambient temperature sensors. If two infrared sensors and two ambient temperature _sensors are used, then T _calc =minimum( T _sky1 ,T _sky2 )-minimum( Tamb1 ,T _amb2 )-more debiasing. The minimum value in the readings from multiple sensors of the same type is used to bias the results towards lower temperature values which would indicate lower cloud cover, and results in higher tone levels to bias the results towards avoiding glare.

In another implementation, the logic of module D' may switch from using the local ambient temperature sensor to use the weather feed, for example, when ambient temperature sensor readings become unavailable or inaccurate, where the ambient temperature sensor is reading heat radiation from a local source, such as from a roof. In this implementation, a temperature value ( T _calc ) is calculated based on the sky temperature reading and the ambient temperature reading ( T _weather ) from the weather feed. In this implementation, the temperature value is calculated as: T _calc = minimum( T _sky1 ,T _{sky2 ,} ...) - T _weather - cloudy bias ( Equation 3 )

In another implementation, the temperature value ( T _calc ) is calculated based on the reading of the difference Δ between the sky temperature and the ambient temperature as measured by two or more IR sensor devices, each IR sensor device having an on-board infrared sensor measurement and ambient temperature sensor. In this implementation, the temperature value is calculated as: T _calc =minimum(Δ ₁ ,Δ ₂ ,...) - cloudy bias ( Equation 4 )

Where Δ ₁ , Δ ₂ . . . are the readings of the difference Δ between the sky temperature and the ambient temperature measured by a plurality of IR sensor devices.

In an implementation using Equation 1 and Equation 3, the control logic uses the difference between the sky temperature and the ambient temperature to determine the filtered IR sensor value input to module D to determine cloud conditions. Ambient temperature readings tend to fluctuate less than sky temperature readings. By using the difference between the sky temperature and the ambient temperature as input to determine the tint state, the determined tint state over time can fluctuate to a lesser extent and provide a more stable tinting of the window.

In another implementation, the control logic calculates T _calc based solely on sky temperature readings from two or more infrared sensors. In this embodiment, the IR sensor value determined by module D' and input into module D is based on the sky temperature reading and not on the ambient temperature reading. In this case, module D determines cloud conditions based on sky temperature readings. While the above implementation for determining T _calc is based on two or more redundant sensors of each type, it should be understood that the control logic could be implemented with readings from a single sensor of a different type.

At operation 2330 , the processor updates the short-term rectangular window and the long-term rectangular window by T _calc determined at operation 2320 . To update the rectangular window, the most recent sensor readings are added to the rectangular window, and the oldest sensor readings are removed from the rectangular window. For module D and other control logic described herein, filtered sensor values are used as input to make shading decisions. Module D' and other logic described herein determine the filtered sensor values using short-term and long-term rectangular windows (filters). Short rectangular windows (e.g., rectangular windows using sample values taken over 10 minutes, 20 minutes, 5 minutes, etc. ) are based on a smaller number of sensor samples (e.g., n =1, 2, 3, . The rectangular window (illumination) values may be based on an average, mean, median or other representative value of the sample values within the rectangular window. In one example, the short rectangular window value is the median value of the sensor samples, and the long rectangular window value is the median value of the sensor samples. Module D' typically uses a rolling median of sensor samples for each of the short and long rectangular window values. In another example, the short rectangular window value is the average value of the sensor samples, and the long rectangular window value is the average value of the sensor samples. Module C1 typically uses filtered light sensor values determined from short and/or long rectangular window values based on the average of sensor samples.

Since the short rectangular window is based on a smaller number of sensor samples, the short rectangular window follows the current sensor reading more closely than the long rectangular window. Thus, short rectangular windows respond faster and to a greater extent to rapidly changing conditions than long rectangular windows. Although both the calculated short and long rectangular window values lag the sensor readings, the short rectangular window value will lag less than the long rectangular window value. Short rectangular windows tend to react faster to current conditions than long rectangular windows. Long rectangular windows can be used to smooth the window controller's response to frequent short duration weather fluctuations (such as passing clouds), while short rectangular windows do not also smooth, but respond more quickly to rapid and significant weather changes (such as overcast conditions). In the case of passing clouds, the control logic using only the long rectangular window value will not respond quickly to the current condition of passing clouds. In this case, the long rectangle window value can be used in the shading decision to determine an appropriate high tone level. In cases where fog dissipates, it may be more appropriate to use short-term rectangular window values in shading decisions. In this case, the short-term rectangular window reacts more quickly to new solar conditions after the fog dissipates. By using short-term rectangular window values to make tinting decisions, tintable windows can quickly adjust to solar conditions while maintaining occupant comfort when fog dissipates quickly.

At operation 2340 , the processor determines a short rectangular window value ( Sboxcar value) and a long rectangular window value ( Lboxcar value) based on the current sensor readings in the rectangular window updated at operation 2330 . In this example, each rectangular window value is calculated by taking the median value of the sensor readings in the rectangular window after the last update performed at operation 2330 . In another implementation, each rectangular window value is calculated by taking the mean of the current sensor readings in each rectangular window. In other implementations, other calculations of sensor readings in each rectangular window may be used.

In certain implementations, the control logic described herein evaluates the difference between the short-term and long-term rectangular window values to determine which rectangular window value to implement when making shading decisions. For example, when the absolute value of the difference between the short-term rectangular window value and the long-term rectangular window value exceeds a threshold value, the short-term rectangular window value can be used in the coloring decision. In this case, the short rectangular window of sensor reading values in the short term is larger than the value of the long term sensor reading by a threshold value, which may indicate short term fluctuations of sufficient significance, eg, a large cloud that may suggest a transition to a lower hue state. If the absolute value of the difference between the short and long rectangular window values does not exceed the threshold value, the long-term rectangular window is used. Returning to FIG. 23 , at operation 2350 the logic evaluates whether the value of the absolute value of the difference between the Sboxcar value and the Lboxcar value is greater than a delta threshold (|Sboxcar value−Lboxcar value|>delta threshold). In some cases, the value of the delta threshold is in the range of 0 millidegrees Celsius to 10 millidegrees Celsius. In one instance, the value of the delta threshold is 0 millidegrees Celsius.

If the absolute value of the difference is above the delta threshold, then the Sboxcar value is assigned to the IR sensor value, and the short-term rectangular window is reset to empty its value (operation 2360 ). If the absolute value of the difference is not higher than the delta threshold, then the Lboxcar value is assigned to the IR sensor value and the long-term rectangular window is reset to empty its value (operation 2370 ). At operation 2380 , the filtered IR sensor values are saved to the IR sensor measurement results database for retrieval by module D. Alternatively, the filtered IR sensor values can be passed directly to module D.

- Example of control logic for making coloring decisions based on infrared sensor and/or light sensor readings depending on am, day, night, night range

In some implementations, the tinting control logic uses filtered values of temperature readings from infrared sensors and ambient temperature readings to determine cloud conditions in the morning and evening and/or at times immediately before sunrise. Since infrared sensors typically operate independently of sunlight intensity levels, this allows the tinting control logic to determine cloud conditions immediately before sunrise, and maintain proper tint levels during the morning and evening as the sun sets. Additionally, readings from the infrared sensor can be used to determine cloud conditions even when the visible light sensor is covered or otherwise blocked. During daylight hours when the infrared sensor is enabled, the tinting control logic determines a first tint level based on the infrared sensor reading and the ambient temperature reading, and determines a second tint level based on the light sensor reading, and then uses the largest of the first and second tint levels. If the IR sensor is not enabled, the control logic uses a second tone level based on the light sensor reading.

In one implementation, the control logic described with respect to FIG. 22 implements module C1 and/or module D depending on whether the calculated time ti _is during the am, day or night range (as determined by sun altitude). An example of this control logic is described in detail with respect to FIG. 24 .

24 illustrates a flowchart 2400 depicting control logic for making shading decisions using infrared sensor and/or light sensor data depending on whether the calculated time ti _is during the morning, day, or night range, according to an implementation. Examples of certain operations of the control logic described with respect to the flow diagram illustrated in FIG. 24 are described with reference to the flow diagrams illustrated in FIGS . 26-28 . In one aspect, this control logic is predictive logic that pre-computes the hue level to which the window should transition. In this aspect, calculations in modules A1, B, C1 and D are performed to determine an appropriate _tone level at a future time (ie, ti = present time plus duration of transition time such as one or more windows). For example, the time used in the calculation may be a time in the future long enough to allow the transition to complete after the hue command is received. In such cases, the window controller can send tint instructions prior to the transition. By the time the transition is complete, one or more windows will have transitioned to the desired tint level for that future time.

In the flowchart 2400 illustrated in FIG. 24 , calculations of the control logic are performed at intervals clocked by the timer at operation 2405 . In one implementation, logical calculations are performed at constant time intervals. In one example, logic calculations are performed every 2 to 5 minutes. In another example, calculations may need to be performed on a less frequent basis, such as every 30 minutes or every 20 minutes, such as a hue transition for a large area electrochromic window that may take 30 minutes or more to transition.

At operation 2412 , the control logic of module A1 is implemented to determine a tint level that takes into account occupant comfort from direct sunlight through the one or more electrochromic windows onto the occupant or their active area. First, the control logic is used to determine whether the sun azimuth angle is outside the critical angle of one or more electrochromic windows. The logic of module A1 is to calculate the position of the sun in the sky based on the latitude and longitude of the building with windows and the time of day t _i and the day of the year (date). The position of the sun includes the solar azimuth (also called solar azimuth). Publicly available programs provide calculations for determining the position of the sun. Critical angles are imported from configuration files for one or more windows. If it is determined that the sun azimuth is outside the critical angle, then it is determined that the sun shines at such an angle that direct sunlight does not enter the room or rooms with the window or windows, and control logic continues to module B at operation 2414 . In this case, module A1 passes the "sunny" tone level (ie, the lowest tone state) to module B as input.

If, on the other hand, it is determined that the azimuth of the sun is between the critical angles of the one or more windows, then the sunlight shines at an angle such that direct sunlight can enter the room through the one or more windows. In this case, the logic of module A1 is implemented to calculate the penetration depth at _time ti based on the calculated position of the sun and window configuration information, including one or more of the position of the window, the size of the window, the orientation of the window (i.e., the direction it is facing), and details of any external shading. The logic of module A1 is then implemented to determine the tint level that will provide occupant comfort for the calculated penetration depth based on the space type of the room by finding in the occupancy lookup table the desired tint level for the space type associated with the window for the calculated penetration depth (e.g., an office with a table near a window, foyer, conference room, etc.) or other data corresponding to different tint levels with the space type and penetration depth. A space type and occupancy lookup table or similar data is provided as input to module A1 from a configuration file associated with one or more windows. In some cases, the tint rating may also be based on providing sufficient natural lighting into a room with one or more windows. In this case, the tonal level determined for the space type and the calculated penetration depth is provided as input to module B.

An example of an occupancy lookup table is provided in FIG. 25 . The values in the table are in terms of hue grades and associated SHGC values in brackets. Figure 25 shows different hue levels (SHGC values) for different combinations of calculated penetration values and space types. The scale is based on eight hue scales, including 0 (brightest), 5, 10, 15, 20, 25, 30, and 35 (brightest). The brightest hue level 0 corresponds to a SHGC value of 0.80, hue level 5 corresponds to a SHGC value of 0.70, hue level 10 corresponds to a SHGC value of 0.60, hue level 15 corresponds to a SHGC value of 0.50, hue level 20 corresponds to a SHGC value of 0.40, hue level 25 corresponds to a SHGC value of 0.30, hue level 30 corresponds to a SHGC value of 0.20, and hue level 35 (darkest) corresponds to a SHGC value of 0.10. The illustrated example includes three space types: Desk 1, Desk 2, and Foyer and six penetration depths.

At operation 2415 , the control logic of module B is implemented to determine a tint level based on the predicted irradiance under clear sky conditions (clear sky irradiance). Module B is used to predict the irradiance at one or more windows under clear sky conditions at t _i and the maximum clear sky irradiance at all times. The maximum clear-sky irradiance is the highest irradiance level predicted for all periods of time for clear-sky conditions. Clear sky irradiance is calculated based on the latitude and longitude coordinates of the building, the window orientation (ie, the direction the window faces) and the time of day t _i and the day of the year. Predicted values of clear sky irradiance can be calculated using open source software (eg, Radiance). Module B typically determines a tone level that is darker than the tone level input from module A1. The tint level judged by Module B transmits less heat than the predicted reference glass would transmit at maximum clear-air irradiance. The logic of module B determines the hue class by stepping up the hue class input from module A1, and selects a hue class where the predicted internal irradiance in the room based on the clear-sky irradiance at ti is less than or equal to the reference internal irradiance, where: internal irradiance=hue class SHGC×clear-sky irradiance, and reference internal _irradiance =reference SHGC×maximum clear-air irradiance. The SHGC of the reference glass is imported into module B from the configuration file. The hue levels from module B are provided as input to modules C1 and D.

Depending on whether time t _i is during the morning, day or night range, the control logic uses infrared sensor and/or light sensor data to make coloring decisions. The control logic determines whether time ti _is during the am, day, evening or night range based on the sun elevation. The logic of module A1 determines the sun position at time t _i , including the sun elevation angle. The sun elevation angle is transmitted from module A1 to modules C1 and D. At operation 2422 , the control logic determines whether the calculated solar elevation angle at _time ti is less than zero. If it is determined that the sun elevation at time ti is less than 0, then it is nighttime and the control logic sets _the nighttime tint state at operation 2424 . An example of a night tone state is the clear tone level, which is the lowest tone state. Clear tint levels can be used as night tint states, for example, to provide security by allowing security personnel outside the building to look through clear windows into the building's interior lit rooms. Another example of a nighttime tint state is the highest tint level, which may also provide privacy and/or security by not allowing others to see inside the building at night when the windows are in the darkest tint state. If it is determined that the sun elevation at time ti is less than 0, then the control logic determines at operation 2490 whether there is a change _in place. If the change is not in place, the final tint level is set to the night tint level. If the modification is in place, the control logic sets the final tint level to the modified value at operation 2492 . At operation 2496 , control logic is implemented to communicate the final tint level to transition the one or more windows to the final tint level. The control logic then proceeds to the timer at operation 2405 for calculation at the next time interval.

If it is determined at operation 2422 that the calculated sun elevation at time t _i is greater than or equal to 0, then the control logic determines at operation 2430 whether the sun elevation is less than a sun elevation threshold. If the sun elevation angle is less than the threshold value of the sun elevation angle, the time t _i is in the morning or evening. In one example, the sun elevation threshold is less than 10 degrees. In another example, the sun elevation threshold is less than 15 degrees. In another example, the sun elevation threshold is less than 20 degrees. If the sun elevation is less than the threshold value of the sun elevation, the control logic determines whether the sun elevation is increasing.

At operation 2432 , the control logic operates to determine whether it is morning based on whether the sun elevation is increasing or decreasing. The control logic determines whether the sun elevation is increasing or decreasing by comparing t _i with a calculated sun elevation value taken at another time. If the control logic determines that the sun elevation is increasing, then it is determined to be AM and the control logic executes the AM IR sensor algorithm implementation of module D at operation 2434 . One example of an AM IR sensor algorithm that may be used is described with respect to flowchart 2600 in FIG. 26 . Module D typically queries an infrared sensor measurement database for filtered IR sensor values at the current time, and determines a cloud condition and associated hue level based on the filtered IR sensor values. If the filtered IR sensor value is below the following threshold, then it is a "sun" condition and the tint level from module D is set to the highest tint level. If the filtered IR sensor value is above an upper threshold, then it is a "cloudy" condition and the tint level from module D is set to the lowest tint level. If the filtered IR sensor value is less than or equal to the upper threshold and greater than or equal to the lower threshold, then the tone level from module D is set to a midtone level. If the control logic determines at operation 2432 that the sun elevation angle does not increase (decrease), then it is determined to be night and the control logic executes the night IR sensor algorithm implementation of module D at operation 2436 . One example of an evening IR sensor algorithm that may be used is described with respect to the flowchart 2700 illustrated in FIG. 27 .

After executing the morning or evening IR sensor algorithm for module D to determine a tint level based on module D, the control logic determines whether there is a change in place at operation 2490 . If the change is not in place, the final tint level is set to the tint level determined by module D. If the modification is in place, control logic sets the final tint level to the modified value at operation 2492 . At operation 2496 , the control logic is implemented to communicate a final tint level to transition one or more electrochromic devices on the one or more windows to the final tint level. The control logic then proceeds to the timer at operation 2405 for calculation at the next time interval.

If it is determined in operation 2430 that the sun elevation is not less than (greater than or equal to) the sun elevation threshold, then time t _i is during the daytime range and the control logic executes a daytime algorithm implementing module C1 and/or module D to determine a hue level based on light sensor and/or infrared sensor readings ( operation 2440 ). The control logic then determines whether a manipulation is in place at operation 2490 . If the change is not in place, the final tint level is set to the tint level determined by module C1 and/or module D's daytime algorithm. One example of a daytime algorithm that may be used is described with respect to the flowchart 2800 illustrated in FIG. 28 . If the modification is in place, the control logic sets the final tint level to the modified value at operation 2492 . At operation 2496 , control logic is implemented to communicate the final tint level to transition the one or more windows to the final tint level. The control logic then proceeds to the timer at operation 2405 for calculation at the next time interval.

In one embodiment, instead of performing the morning IR sensor algorithm for module D at operation 2434 , the night IR sensor algorithm for module D at operation 2436 , and the daytime algorithm for module C1 and/or module D at operation 2440 , at operation 2434 the morning light sensor algorithm for module C1 is used, at operation 2436 the night light sensor algorithm for module C1 is used, and at operation 2440, module C is used 1 daytime algorithm.

-Example of morning IR sensor algorithm and evening IR sensor algorithm of module D

Module D queries the IR sensor measurement database for a filtered IR sensor value (or receives the value directly from another logic module), and then determines a cloud condition and associated hue level based on the filtered IR sensor value. If the filtered IR sensor value is below the following threshold, then it is a "sun" condition and the tint level from module D is set to the highest tint level. If the filtered IR sensor value is above an upper threshold, then it is a "cloudy" condition and the tint level from module D is set to the lowest tint level. If the filtered IR sensor value is less than or equal to the upper threshold and greater than or equal to the lower threshold, then the tone level from module D is set to a midtone level. The upper and lower thresholds used in these calculations are based on whether the morning IR sensor algorithm, the evening IR sensor algorithm, or the daytime algorithm is being implemented.

29 shows a graph of filtered IR sensor values in milliCelsius versus time during a 24 hour period. The graph shows three regions of the range of filtered IR sensor values. The upper area above the upper threshold is "cloudy". Filtered IR sensor values above the upper threshold are in "cloudy" regions. The intermediate areas between the upper and lower thresholds are "intermittently cloudy" or "partly cloudy". The lower region below the lower threshold is the "clear" region also known as the "sun" region. Filtered IR sensor values below the upper threshold are in the "clear" or "sun" region. The graph has two curves of calculated filtered IR sensor values based on readings taken over two 24 hour periods. The first curve 2930 shows calculated filtered IR sensor values taken during the first day with cloudy afternoons. The second curve 2932 shows the calculated filtered IR sensor values taken during the second full sun/clear day. The lower threshold describes the lower boundary between the middle region and the lower region. The upper threshold describes the upper boundary between the middle zone and the upper zone. The lower and upper thresholds used during the evening (evening lower threshold and evening upper threshold) are generally higher than the lower and upper thresholds used during the morning (morning lower threshold and morning upper threshold).

FIG. 26 illustrates a flowchart 2600 depicting the control logic implemented by the AM IR sensor algorithm of module D. The am IR sensor algorithm may be implemented when the tinting control logic determines that the current time is during the am range. The AM IR sensor algorithm is one example of control logic that may be implemented at operation 2434 of the flowchart shown in FIG. 24 when the control logic determines that the sun elevation angle is less than an elevation threshold and the sun elevation angle is increasing.

The control logic of flowchart 2600 begins at operation 2610 and compares the filtered IR sensor value to the AM lower threshold to determine whether the filtered IR sensor value is less than the AM lower threshold. The control logic of module D queries the IR sensor measurement results database or other database to retrieve filtered IR sensor values. Alternatively, the control logic calculates filtered IR sensor values. An example of control logic that may be used to calculate filtered IR sensor values and store the values to an infrared sensor measurement database is the control logic of module D' described with reference to the flowchart in FIG. 23 . The AM lower threshold value is the temperature value at the lower boundary between the lower zone ("sun" or "clear" zone) and the middle zone ("partly cloudy zone") the filtered IR sensor value applied during the AM range. In some implementations, the AM lower threshold is in the range of -20 millidegrees Celsius and 20 millidegrees Celsius. In one example, the morning lower threshold is 1 degree Celsius.

If it is determined at operation 2610 that the filtered IR sensor value is less than the morning lower threshold, then it is determined that the filtered IR sensor value is in the lower region of the "sunny" or "sun" region. In this case, the control logic sets the tint level from module D to a high tint state (eg, tint level 4) and passes the tint level from module D ( operation 2620 ).

If it is determined at operation 2610 that the filtered IR sensor value is not less than the lower AM threshold, then control logic proceeds to determine at operation 2630 whether the filtered IR sensor value is less than or equal to the upper AM threshold and greater than or equal to the lower AM threshold. The AM upper threshold value is the temperature at the upper boundary of filtered IR sensor values between the middle zone ("partly cloudy zone") and the upper zone ("cloudy" zone) applied during the morning range of the day. In some implementations, the upper AM threshold is in the range of -20 millidegrees Celsius and 20 millidegrees Celsius. In one example, the upper morning threshold is 3 millidegrees Celsius.

If it is determined at operation 2630 that the filtered IR sensor value is less than or equal to the upper AM threshold and greater than or equal to the lower AM threshold, then the filtered IR sensor value is determined to be in an intermediate region that is a "partly cloudy" region ( operation 2640 ). In this case, the control logic sets the tone level of module D to a midtone state (eg, tone level 2 or 3) and passes the tone level from module D.

If it is determined at operation 2630 that the filtered IR sensor value is not less than or equal to the upper AM threshold and greater than or equal to the lower AM threshold (i.e., the filtered sensor value is greater than the upper AM threshold), then the filtered IR sensor value is determined to be in an upper region that is "cloudy" ( operation 2650 ). In this case, the control logic sets the tint level of module D to a low tint state (eg, tint level 2 or a lower tint level) and passes the tint level from module D.

FIG. 27 illustrates a flowchart 2700 depicting the control logic implemented by the night IR sensor algorithm for module D. When the tinting control logic determines that the current time is during the evening range, the evening IR sensor algorithm may be implemented. The evening IR sensor algorithm is one example of control logic that may be implemented at operation 2436 of the flowchart shown in FIG. 24 when the control logic determines that the sun elevation angle is less than an elevation threshold and that the sun elevation angle is decreasing.

The control logic of flowchart 2700 begins at operation 2710 and compares the filtered IR sensor value to the night lower threshold value to determine whether the filtered IR sensor value is less than the night lower threshold value. The control logic of module D queries the IR sensor measurement results database or other database to retrieve filtered IR sensor values. Alternatively, the control logic calculates filtered IR sensor values. An example of control logic that may be used to calculate filtered IR sensor values and store the values to an infrared sensor measurement database is the control logic of module D' described with reference to the flowchart in FIG. 23 . The evening lower threshold is the temperature value at the lower boundary between the lower zone ("clear" or "clear" zone) and the middle zone ("partly cloudy zone") the filtered IR sensor value applied during the evening range. In some implementations, the nighttime lower threshold is in the range of -20 degrees Celsius and 20 degrees Celsius. In one example, the lower threshold at night is 2 millidegrees Celsius.

If it is determined at operation 2710 that the filtered IR sensor value is less than the evening lower threshold, then the filtered IR sensor value is in the lower region of the "sunny" or "sun" region. In this case, the control logic sets the tint level from module D to a high tint state (eg, tint level 4) and passes the tint level from module D at operation 2720 .

If it is determined at operation 2710 that the filtered IR sensor value is not less than the lower night threshold, then control logic proceeds to determine at operation 2730 whether the filtered IR sensor value is less than or equal to the upper night threshold and greater than or equal to the lower night threshold. The evening upper threshold value is the temperature at the upper boundary of filtered IR sensor values between the middle zone ("partly cloudy zone") and the upper zone ("cloudy" zone) applied during the night range of the day. In some implementations, the nighttime upper threshold is in the range of -20 degrees Celsius and 20 degrees Celsius. In one example, the upper threshold at night is 5 millidegrees Celsius.

If it is determined at operation 2730 that the filtered IR sensor value is less than or equal to the upper evening threshold and greater than or equal to the lower evening threshold, then the filtered IR sensor value is determined to be in an intermediate region of the "partly cloudy" region ( operation 2740 ). In this case, the control logic sets the tone level of module D to a midtone state (eg, tone level 2 or 3) and passes the tone level from module D.

If it is determined at operation 2730 that the filtered IR sensor value is not less than or equal to the upper evening threshold and greater than or equal to the lower evening threshold (i.e., the filtered sensor value is greater than the upper evening threshold), then the filtered IR sensor value is determined to be in the upper region of "cloudy" ( operation 2750 ). In this case, the control logic sets the tint level of module D to a low tint state (eg, tint level 2 or a lower tint level) and passes this tint level from module D.

-Example of daytime algorithm for module C1 and/or module D

During the day, the temperature readings taken by the infrared sensor may tend to fluctuate if the localized area around the infrared sensor is heated. For example, an infrared sensor on a roof can Heated by the roof as it absorbs heat from the midday sun. In some implementations, the daytime algorithm disables the use of IR sensor readings in its tinting decisions under certain circumstances, and uses module Cl to determine hue levels solely from light sensor readings. In other cases, the daytime algorithm uses module D to determine a first tint level based on IR sensor readings, uses module C1 to determine a second tint level based on light sensor readings, and then sets the tint level to the maximum of the first and second tint levels.

FIG. 28 illustrates a flowchart 2800 depicting control logic that may implement the daytime IR sensor algorithm of module C1 and/or the daylight algorithm of module D's daytime light sensor algorithm. The daytime algorithm is used when the shading control logic determines that the current time is during the daytime range. The daytime algorithm is one example of control logic that may be implemented at operation 2440 of the flowchart shown in FIG. 24 when the sun elevation angle is greater than or equal to 0 and less than or equal to the elevation threshold.

At operation 2810 , it is determined whether to enable use of IR sensor readings. In one case, the default setting for the tinting control logic is to disable the use of IR sensor readings unless light sensor readings are not available, for example due to a malfunctioning light sensor. In another case, the control logic disables use of the IR sensor readings if IR sensor data is not available, eg, due to a malfunctioning IR sensor. If at operation 2810 it is determined that use of IR sensor readings is enabled, then the control logic executes both the daytime IR sensor algorithm for module D and the daytime light sensor algorithm for module C1 ( operation 2820 ). If it is determined at operation 2810 that use of IR sensor readings is not enabled, then the control logic executes the daylight sensor algorithm for module C1 ( operation 2850 ).

At operation 2830 , the logic of the daytime IR sensor algorithm of module D is executed to determine the first hue state. The filtered IR sensor values are retrieved from the IR sensor measurement results database or other databases. Instead, the logic of the daytime IR sensor algorithm calculates filtered IR sensor values. One example of logic that may be used to calculate filtered IR sensor values and store the values to the IR sensor measurement database is the control logic of module D' described with reference to the flowchart in FIG. 23 . The logic of the daytime IR sensor algorithm compares the filtered IR sensor value to the lower daytime threshold to determine whether the filtered IR sensor value is less than the lower daytime threshold, greater than the upper daytime threshold, or between the lower daytime threshold and the upper daytime threshold. The lower daytime threshold is the temperature at the lower boundary between the lower zone ("sun" or "clear" zone) and the middle zone ("partly cloudy" zone) the filtered IR sensor value applied during the daytime range. In some implementations, the daytime lower threshold is in the range of -20 degrees Celsius and 20 degrees Celsius. In one example, the daytime lower threshold is -1 millidegree Celsius. The daytime upper threshold value is the temperature value at the upper boundary of filtered IR sensor values between the middle zone ("partly cloudy zone") and the upper zone ("cloudy" zone) applied during the night range of the day. In some implementations, the daytime upper threshold is in the range of -20 degrees Celsius and 20 degrees Celsius. In one example, the daytime upper threshold is 5 millidegrees Celsius. If it is determined that the filtered IR sensor value is less than the lower daytime threshold value, then the filtered IR sensor value is in a region lower than the "sunny" or "sun" region. In this case, the control logic sets the first tint level from module D to a high tint state (eg, tint level 4). If the filtered IR sensor value is determined to be less than or equal to the daytime upper threshold and greater than or equal to the upper daytime threshold, then the filtered IR sensor value is determined to be in the middle region which is a "partly cloudy" region. In this case, the control logic sets the first tone level to a mid-tone state (eg, tone level 2 or 3). If it is determined that the filtered IR sensor value is not less than or equal to the upper daytime threshold and greater than or equal to the lower daytime threshold (i.e., the filtered sensor value is greater than the upper daytime threshold), then the filtered IR sensor value is determined to be in the upper region of the "cloudy" region. In this case, the control logic sets the first tone level of module D to a low tone state (eg, tone level 2 or lower).

In operation 2832 , the logic of the daylight sensor algorithm of module C1 is executed to determine the second tone level. Module C1 uses the light sensor readings to determine the second hue level based on the instantaneous irradiance. One example of control logic for module Cl that may be used to determine the second tone level is described in the following sections with respect to the flowchart 3000 shown in FIG. 30 .

At operation 2840 , the logic of the daytime algorithm calculates the maximum of a first hue level using module D based on IR sensor readings and a second hue level using module C1 based on light sensor readings. The tint level from the daytime algorithm is set to the maximum of the calculated first tint state based on IR sensor readings and the calculated second tint level based on light sensor readings. Returns the hue level from module C1 or D.

If it is determined at operation 2810 that use of IR sensor readings is not enabled, then the control logic executes the daylight sensor algorithm for module C1 ( operation 2850 ). In operation 2850 , the logic of the daylight sensor algorithm of module C1 is executed to determine the second tone level. In this case, the tint state from the daytime algorithm is set to the second tint level based on the light sensor reading, and this tint level from module C1 is returned. An example of the control logic of module C1 that may be used to determine the second tone level is described with respect to the flowchart shown in FIG. 30 .

- Example of module C1

As shown, FIG. 30 includes a flowchart 3000 depicting the control logic of an example of module C1 for determining the tint level of one or more electrochromic windows according to one aspect. Module C1 receives hue levels from module B as input.

At operation 3020 , current light sensor values reflecting conditions outside the building are received, and qualification is implemented to calculate a suggested tint level to apply. In one example, the current light sensor value is the maximum value of measurement results obtained by multiple light sensors (eg, 13 light sensors in a multi-sensor device) at one sample time. In another example, the light sensor value is a filtered rolling average of multiple readings taken at different sample times, where each reading is the maximum value of measurements taken by the multiple light sensors. One example of control logic that may be used to calculate the current light sensor value is depicted in flow diagram 3100 in FIG. 31 , which depicts the control logic for module C1 '.

Returning to FIG. 30 , at operation 3020 , thresholds are used to calculate suggested hue levels by determining whether the current filtered photosensor value has crossed one or more thresholds over a period of time. For example, the time period may be the time period between the current time and the time of the last sample taken by the light sensor or between the current time and the first reading of a plurality of previously taken sample readings. Light sensor readings may be taken periodically, such as once a minute, every 10 seconds, every 10 minutes, and the like. In one implementation, the qualification uses two thresholds: a lower light sensor threshold and an upper light sensor threshold. If it is determined that the light sensor value is above the upper light sensor threshold value, then the light sensor value is in the higher region of the "sunny" or "sun" region. In this case, the control logic determines the suggested tone level from module C1 as a high tone state (eg, tone level 4). If the light sensor value is determined to be less than or equal to the upper light sensor threshold and greater than or equal to the lower light sensor threshold, then the light sensor value is determined to be in the middle of the "partly cloudy" area. In this case, the control logic determines that the suggested tone level from module C1 is a mid-tone state (eg, tone level 2 or 3). If the light sensor value is determined to be greater than the evening upper threshold, then the light sensor value is determined to be in the upper region of the "cloudy" region. In this case, the control logic determines that the suggested tone level from module C1 is a low tone state (eg, tone level 2 or lower).

If the current time is after the lockout period has ended, control logic calculates a suggested tint level at operation 3020 based on the conditions monitored during the lockout period. The suggested tone level calculated based on the conditions monitored during the lockout period is based on a statistical evaluation of the monitored input. Various techniques are available for statistical evaluation of the inputs monitored during the latency period. One example is the hue level averaged during the waiting time. During the wait time, the control logic performs operations of monitoring the input and calculating a hue level, eg, determined using one or more of modules A1, B, and C1. The operation then averages the determined hue levels over the wait time to determine which direction is suggested for a hue region transition.

In operation 3025 , it is determined whether the current time is during a lock period. If the current time is during the lockout period, then module C1 does not change the tone level received from module B. During the lockout period, the light sensor value is monitored for external conditions. Additionally, the control logic monitors the suggested hue level determined by operation 3020 during the lockout period. If the current time is determined not to be during the lockout period, then the control logic proceeds to operation 3030 .

At operation 3030 , the logic of module C1 continues to determine whether the current information suggests a tone shift. This operation 3030 compares the suggested tint level determined at operation 3020 with the current tint level applied to one or more windows to determine if the tint level is different. If the suggested tint level is not different from the current tint level, the tint level is not changed.

In operation 3050 , if the suggested hue level is different from the current hue level, module C1 sets a new hue level, which is one tone level towards the suggested hue level determined in operation 3020 (even if the suggested hue level is two or more hue levels from the current hue level). For example, if the suggested tone range determined in operation 3020 is from the first tone level to the third tone level, then the tone level returned by module C1 shifts one tone level to the second tone level.

At operation 3070 , a lockout period is set such that transitions to other tone levels are locked out during the lockout period. During the lockout period, the light sensor value is monitored for external conditions. Additionally, the control logic calculates during intervals a suggested hue region based on the conditions monitored during the lockout period. The new tone level delivered from module C1 is determined in operation 3050 as one towards the suggested tone level determined in operation 3020 .

- Example of module C1'

FIG. 31 illustrates a flowchart 3100 depicting the logic of module C1' according to certain implementations. The logic of the module C1' can be executed by one or more processors of the local window controller, the network controller, the main controller or the multi-sensor device. At operation 3110 , the processor performing the operations of module C1' receives as input sensor readings at the current time. Light sensor readings may be received via a communication network at the building, for example, from a rooftop multi-sensor installation. The received light sensor readings are instant irradiance readings.

In one implementation, the logic of module C1' receives and uses raw light sensor readings from measurements taken by two or more light sensors at the building (eg, of a rooftop multi-sensor installation). Two or more light sensors are usually used to provide redundancy. According to one aspect, the logic of module C1' uses raw light sensor readings from measurements taken by two light sensor devices at the building. In another aspect, the logic of module C1' uses raw light sensor readings from measurements taken by 1 to 10 light sensors at the building. In another aspect, the logic of module C1' uses raw light sensor readings from measurements taken by thirteen (13) light sensors at the building.

At operation 3120 , light sensor values are calculated based on raw measurements taken by the two or more light sensors. For example, the light sensor value can be calculated as the maximum value of measurements taken by two or more light sensors at a single sample time.

In operation 3130 , the processor updates the short-term rectangular window and the long-term rectangular window with the light sensor value determined in operation 3120 . In module C1' and other control logic described herein, the filtered light sensor values are used as input to make shading decisions. Module C1' and other logic described herein determine the filtered sensor values using short-term and long-term rectangular windows (filters). Short rectangular windows (e.g., rectangular windows using sample values taken over 10 minutes, 20 minutes, 5 minutes, etc.) are based on a smaller number of sensor samples (e.g., n=1, 2, 3, . The rectangular window (illumination) values may be based on an average, mean, median or other representative value of the sample values within the rectangular window. In one example, the short rectangular window values are the average of the sensor samples, and the long rectangular window values are the average of the light sensor samples. Module D' typically uses a rolling average of the sensor samples for each of the short and long rectangular window values. In another example, the short rectangular window value is the average value of the sensor samples, and the long rectangular window value is the average value of the sensor samples.

At operation 3140 , the processor determines a short rectangular window value ( Sboxcar value) and a long rectangular window value ( Lboxcar value) based on the current light sensor readings in the rectangular window updated at operation 3130 . In this example, each rectangular window value is calculated by taking the average of the light sensor readings in the rectangular window after the last update performed at operation 3130 . In another example, each rectangular window value is calculated by taking the median value of the light sensor readings in the rectangular window after the last update performed at operation 3130 .

At operation 3150 , the logic evaluates whether the absolute value of the difference between the Sboxcar value and the Lboxcar value is greater than a delta threshold (|Sboxcar value−Lboxcar value|>delta threshold). In some cases, the value of the delta threshold is in the range of 0 millidegrees Celsius to 10 millidegrees Celsius. In one instance, the value of the delta threshold is 0 millidegrees Celsius.

If the difference is above the delta threshold, then the Sboxcar value is assigned to the light sensor value and the short-term rectangular window is reset to empty its value (operation 3160 ). If the difference is not higher than the delta threshold, then the Lboxcar value is assigned to the light sensor value and the long-term rectangular window is reset to empty its value (operation 3170 ). At operation 3180 , the light sensor values are saved to a database.

While a single infrared sensor is described as being included in some implementations of the infrared cloud detector, according to another implementation, two or more infrared sensors are available for redundancy should one infrared sensor fail and/or be obscured by, for example, bird droppings or another environmental agent. In one aspect, two or more infrared sensors may be included, facing different orientations to capture infrared radiation from different fields of view and/or at different distances from the building/structure. If two or more infrared sensors are located within the housing of the infrared cloud detector, the infrared sensors are typically offset from each other by a distance sufficient to reduce the likelihood that an obscuring object will affect all the infrared sensors. For example, the infrared sensors can be separated by at least about one inch or at least about two inches.

In some embodiments described herein, the control logic determines the tone level based on conditions that are likely to occur at a future time (also referred to herein as "future conditions"). For example, a tint level may be determined _based on the likelihood of occurrence of cloud conditions at a future time (eg, ti = present time + duration such as transition time of one or more electrochromic windows). The future time used in these logic operations can be set to a time in the future sufficient to allow the transition of the window to the tint level to be completed after the control command is received. In such cases, the controller can send the command at the current time before the actual transition. By the time the transition is complete, the window will have transitioned to the desired tint level for that future time. In other embodiments, the disclosed control logic can be used to determine tone levels based on conditions that occur or are likely to occur at the current time, for example, by setting the duration to zero. For example, in some electrochromic windows, the transition time to a new tint level (eg, to a mid-tone level) may be so short that it would be appropriate to send a command to transition to a tint level based on the current time.

It should be understood that the present invention as described above can be implemented in the form of control logic using computer software in a modularized or integrated manner. Based on the present disclosure and the teachings provided herein, one of ordinary skill in the art will know and appreciate other ways and/or methods of implementing the present invention using hardware and combinations of hardware and software.

Any of the software components or functions described in this application may be implemented as software code to be executed by a processor using, for example, conventional or object-oriented techniques, using any suitable computer language such as Java, C++ or Perl. Software code may be stored as a series of instructions or commands on a computer readable medium such as random access memory (RAM), read only memory (ROM), magnetic media such as a hard disk or floppy disk, or optical media such as CDROM. Any such computer-readable media may reside on or within a single computing device, and may reside on or within different computing devices within a system or network.

While the foregoing disclosed embodiments have been described in some detail to facilitate understanding, the described embodiments are to be considered illustrative rather than restrictive. It will be apparent to those of ordinary skill in the art that certain changes and modifications may be practiced within the scope of the appended claims.

One or more features from any embodiment may be combined with one or more features from any other embodiment without departing from the scope of the present disclosure. In addition, modifications, additions or omissions to any of the embodiments may be made without departing from the scope of the present disclosure. Components of any embodiment may be integrated or separated according to particular needs without departing from the scope of the present disclosure.

300‧‧‧Infrared Cloud Detector System

310‧‧‧Infrared Cloud Detector

312‧‧‧Shell

314‧‧‧Infrared sensor

315‧‧‧Conical Field of View

316‧‧‧Ambient temperature sensor

320‧‧‧External Visible Light Sensor

Room 330‧‧‧

332‧‧‧tintable windows

334‧‧‧table

340‧‧‧Controller

Claims (26)

A controller for controlling the tint of one or more tintable windows of a building, the controller comprising: a computer readable medium having a tint level configured to determine a tint level for the one or more tintable windows based on a cloud condition; a processor in communication with the computer readable medium and in communication with a local window controller of the one or more tintable windows, wherein the processor is configured to be based, in part, on: (A) light sensor readings from at least one light sensor and infrared sensor readings from at least one infrared sensor one or both of these and (B) a future time during a morning range, a daytime range, or an evening range to determine the cloud condition; calculate the tint level for the one or more tintable windows based on the determined cloud condition; and send a tint command to the local window controller to transition the tint of the one or more tintable windows to the calculated tint level. The controller according to claim 1, wherein said cloud condition is predicted to occur at said future time. The controller of claim 1, wherein the processor is further configured to determine whether the future time is during the morning range, the daytime range, or the evening range based on a sun elevation angle at the future time. The controller according to claim 1, wherein determining the cloud condition is based in part on an average value, an average value or a median value of the light sensors and/or an average value, an average value or a median value of the infrared sensors. The controller as described in item 1 of the scope of the patent application, wherein the processor is further further configuring to: (i) determine the cloud condition based in part on the infrared sensor reading if the future time is during the morning range or during the evening range, and (ii) determine the cloud condition based on the light sensor reading if the future time is during the day range. The controller of claim 1 wherein determining said cloud condition is based in part on a minimum of said infrared sensor readings if said future time is during said am range or during said evening range. The controller of claim 1 wherein determining said cloud condition is based in part on a calculated difference between a minimum of said infrared sensor readings and a minimum of ambient temperature sensor readings if said future time is during said morning range or during said evening range. The controller of claim 1 wherein the cloud condition is determined based on a calculated difference between a minimum of the infrared sensor readings and an ambient temperature from a weather feed if the future time is during the morning range or during the evening range. The controller of claim 1, wherein the control logic is configured to determine the cloud condition based in part on one or more of the light sensor reading, the infrared sensor reading, and ambient temperature sensor reading. The controller according to claim 1, wherein the processor is further configured to: determine a first color tone level based on the light sensor reading; determine a second color tone level based on the infrared sensor reading; and calculate the color tone level as a maximum value of the determined first and second color tone levels. The controller as described in item 1 of the scope of the patent application, wherein the processor is configured If the future time is during nighttime, the hue level is calculated as a nighttime tone level. A method, performed by a controller, of controlling the tint of one or more tintable windows of a building, the method comprising: determining a cloud condition based in part on: (A) one or both of light sensor readings from at least one light sensor and infrared sensor readings from at least one infrared sensor and (B) a future time during a morning range, a daytime range, or an evening range; calculating a tint level for the one or more tintable windows based on the determined cloud condition; The tint of the tintable window is converted to the calculated tint scale. The method of claim 12, wherein said cloud condition is predicted to occur at said future time. The method according to claim 12, further comprising: calculating a sun elevation angle; and determining whether the future time is during the daytime range, the morning range or the evening range based on the calculated sun elevation angle. The method according to claim 12, wherein: if the future time is during the morning range or the night range, then determining the cloud condition based partly on the infrared sensor reading; if the future time is during the daytime range, then determining the cloud condition based partly on one or both of the light sensor reading and the infrared sensor reading; and if the future time is during the night time, determining the hue level as a nighttime hue level. The method of claim 12, further comprising calculating a transition time based on a current time and a representative window of a window zone comprising the one or more tintable windows Calculate said future time. The method according to claim 12, further comprising wherein if the future time is within the daytime range: determining a first hue level based on the light sensor reading; determining a second hue level based on the infrared sensor reading; and calculating the hue level as a maximum value of the first and second hue levels. The method of claim 12, further comprising wherein if the future time is during the daytime range and the at least one infrared sensor is disabled, then calculating the hue level from the cloud conditions determined based in part on the light sensor readings. The method according to claim 12, further comprising: determining the cloud condition based on the infrared sensor reading if the future time is within the morning range or the evening range. The method of claim 19, further comprising determining the cloud condition based in part on the light sensor readings if the future time is within the am range. The method of claim 12, further comprising determining said cloud condition based in part on a minimum of said infrared sensor readings if said future time falls during said morning range or said evening range. The method of claim 12, further comprising determining said cloud condition based in part on a calculated difference between a minimum of said infrared sensor readings and a minimum of ambient temperature sensor readings if said future time falls during said am range or said evening range. The method of claim 12, further comprising, based in part on the infrared sensor if the future time is during the morning range or the evening range The cloud condition is determined from a calculated difference between a minimum of readings and an ambient temperature from a weather feed. The method according to claim 12, further comprising determining that the tone level is a night tone level if the future time is during a night time period. A non-transitory computer-readable medium for controlling one or more tintable windows that, when read by one or more processors operatively coupled to said at least one infrared sensor, to said at least one light sensor, and/or to at least one ambient temperature sensor, causes said one or more processors to execute or direct execution of the method of any one of claims 12 to 24. A system for controlling the tint of one or more tintable windows, the system comprising a network configured to: (a) be operatively coupled to the one or more tintable windows and to at least one infrared sensor, at least one light sensor, and at least one ambient temperature sensor; and (b) transmit signals associated with the method of any one of claims 12 to 24.

Images