WO2020192210A1

WO2020192210A1 - A method and a system for managing indoor air quality

Info

Publication number: WO2020192210A1
Application number: PCT/CN2019/128116
Authority: WO
Inventors: Kam Fu CHOW; Pei Jie HUANG; Ling Chun Byron NG
Original assignee: Woofaa Company Limited
Priority date: 2019-03-26
Filing date: 2019-12-25
Publication date: 2020-10-01

Abstract

A method for managing indoor air quality in an area comprises the steps of: determining occupancy of an indoor area（102）; determining a ventilation demand in the indoor area associated with the determined occupancy（106）; and controlling an HVAC unit with an operation profile so as to satisfy the determined ventilation demand（108）. A system for managing indoor air quality in an area is also provided.

Description

A METHOD AND A SYSTEM FOR MANAGING INDOOR AIR QUALITY

TECHNICAL FIELD

The present invention relates to a method and a system for managing indoor air quality, and particularly, although not exclusively, to a system for managing indoor air quality based on indoor occupancy determination.

BACKGROUND

A heating, ventilation, and air conditioning (HVAC) system may be used to control air quality indoor environments, such as apartment buildings, hotel rooms, office premises and vehicles, so as to maintain healthy air conditions including temperature and humidity, as well as freshness of air.

There are various methods to control an HVAC system, for example by using a timer-based air handling unit (AHU) controller. Using this control method, the AHUs periodically supply air to different areas in an area at a constant air flow rate of full-load condition.

SUMMARY OF THE INVENTION

To solve or reduce at least some of the drawbacks associated with the conventional methods for managing indoor air quality, the present invention discloses a method employing a vision artificial intelligent (AI) technology.

In accordance with a first aspect of the present invention, there is provided a method for managing indoor air quality in an area, comprising the steps of: determining an occupancy of an indoor area; determining a ventilation demand in the indoor area associated with the determined occupancy; and controlling an HVAC unit with an operation profile so as to satisfy the determined ventilation demand.

In an embodiment of the first aspect, the step of determining the occupancy includes determining number of people in the indoor area.

In an embodiment of the first aspect, the step of determining number of people comprises the steps of: capturing one or more images of the indoor area including individuals in the indoor area; and identifying the individuals in the images so as to determine the number of people in the indoor area.

In an embodiment of the first aspect, the determination of the number of people is based on visual skeleton recognition of the individuals in the images.

In an embodiment of the first aspect, the method further comprises the steps of: determining a concentration of carbon dioxide in the indoor area; and controlling the HVAC unit based on the determined concentration of carbon dioxide.

In an embodiment of the first aspect, the determination of the number of people is based on face-detection of the individuals in the images.

In an embodiment of the first aspect, the method further comprises the step of providing face-detection and visual skeleton recognition results to a server through wireless connections.

In an embodiment of the first aspect, the method further comprises the step of providing carbon dioxide concentration results to the server through wireless connections.

In an embodiment of the first aspect, the step of controlling the HVAC unit includes controlling the speed of fans of the HVAC unit based on the determined occupancy and the determined concentration of carbon dioxide.

In an embodiment of the first aspect, the method further comprises the step of reporting determined visual recognition results and carbon dioxide results on an external device for user interaction.

In an embodiment of the first aspect, the method further comprises the step of: detecting movement of doors in the indoor area; and determining an effect of indoor air quality due to air exchanges between the indoor area and outdoor environment separated by the doors.

In an embodiment of the first aspect, the method further comprises the steps of: aggregating and analysing the determined visual data, carbon dioxide concentration data and air quality status on the server; and displaying the analysed data on an external device for user interaction.

In an embodiment of the first aspect, the method further comprises the step of generating a heat map based on the analysed data.

In accordance with a second aspect of the present invention, there is provided a system for managing indoor air quality in an area, comprising: a detection module arranged to determine an occupancy of an indoor area; a processing module arranged to determine a ventilation demand in the indoor area associated with the determined occupancy; and a controller module arranged to control an HVAC unit with an operation profile so as to satisfy the determined ventilation demand.

In an embodiment of the second aspect, the detection module includes a face-detection camera.

In an embodiment of the second aspect, the detection module includes a skeleton recognition camera.

In an embodiment of the second aspect, the system further comprises a gas sensor arranged to determine a concentration of carbon dioxide in the indoor area, wherein the controller module is further arranged to control the HVAC unit based on the determined concentration of carbon dioxide.

In an embodiment of the second aspect, the face-detection camera is arranged to provide face-detection results to a server through wireless connections.

In an embodiment of the second aspect, the skeleton recognition camera is arranged to provide visual skeleton recognition results to the server through wireless connections.

In an embodiment of the second aspect, the gas sensor is further arranged to provide carbon dioxide concentration results to the server through wireless connections.

In an embodiment of the second aspect, the controller module includes a variable speed drive (VSD) controller arranged to control the speed of fans of the HVAC unit based on the determined occupancy and the determined concentration of carbon dioxide.

In an embodiment of the second aspect, the detection module includes a door switch arranged to detect movement of doors in the indoor area.

In an embodiment of the second aspect, the server is arranged to determine an effect of indoor air quality due to air exchanges between the indoor area and outdoor environment separated by the doors.

In an embodiment of the second aspect, the server is further arranged to perform data aggregation and analyse the obtained visual data, carbon dioxide concentration data and air quality status, and display the analysed data on an external device for user interaction.

In an embodiment of the second aspect, the server is further arranged to generate a heat map based on the analysed data.

In accordance with a third aspect of the present invention, there is provided a system for managing indoor air quality in an area including multiple zones, each of the zones requires an individually managed air ventilation suitable for a dynamic occupancy of the indoor area, the system comprising: a plurality of visual-based people counting modules arranged to count a number of people in each of the zones in the indoor area, each of the visual-based people counting modules includes a first camera for identifying individuals based on face-detection and a second camera for identifying individuals based on skeleton recognition, the first and the second cameras are IP cameras that connect to a server (not shown) through wireless connections; a plurality of indoor air quality sensors installed in each of the zones in the indoor area arranged to detect the air quality including a concentration of carbon dioxide in each of the zones in the indoor area, and to provide the detection results to the server; a plurality of variable speed drive (VSD) controllers arranged to drive fans in a plurality of HVAC units with different operation profiles for providing fresh air and circulated air to each of the zones in the indoor area based on different air ventilation requirements, wherein the VSD controllers in each zone control the speed of the fans based on the occupancy determined based on the number of people counted by the visual-based people counting module and the detected indoor air quality in a particular zone; a plurality of door switches arranged to detect doors in the indoor area being opened and closed, the server then determines an effect of indoor air quality due to air exchanges between the indoor area and outdoor environment separated by the doors; and a remote management platform that allows a user to interact with the server that performs data aggregation and analytics of visual data obtained from the visual-based people counting module and air quality status determined by the indoor air quality sensors; the remote management platform also allows the user to remotely control each of the HVAC units in different zones in the indoor area through the server and a cloud-based platform, and to customise the operation profiles in each of the individual zones associated with the occupancy of the indoor area.

BRIEF DESCRIPTION OF THE DRAWINGS

Embodiments of the present invention will now be described, by way of example, with reference to the accompanying drawings in which:

Figure 1 is flow chart illustrating a method for managing indoor air quality in an area in accordance with one embodiment of the present invention;

Figure 2 is an image obtained using visual skeleton recognition in accordance with one example embodiment of the present invention;

Figure 3A is a picture illustrating the control of an HVAC unit with a first operation profile in accordance with one example embodiment of the present invention;

Figure 3B is a picture illustrating the control of the HVAC unit in Figure 3A with a second operation profile in accordance with one embodiment of the present invention;

Figure 4 is an image illustrating a reporting dashboard on an external device in accordance with one example embodiment of the present invention;

Figure 5A is a heat map of an area in accordance with one example embodiment of the present invention;

Figure 5B is a heat map of an area in accordance with one example embodiment of the present invention; and

Figure 6 is an image illustrating a system for managing indoor air quality in an area in accordance with one example embodiment of the present invention.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENT

Without wishing to be bound by theories, the inventors have devised that a timer-based air handling unit (AHU) control in a HVAC system introduces deficiency as the occupancy may not be constant at the same time of a day. Dynamic load changes over weekday-weekend may also occur. Therefore, such timer-based control is not energy efficient.

In an alternative example, a demand controlled ventilation (DCV) method in HVAC technology may be used to better control and regulate indoor air quality (IAQ) and with enhanced energy efficiency. For example, DCV may involve determining a carbon dioxide level of an indoor area, and then adjusting the HVAC units to increase the volume exchange of fresh air or outside air into the indoor area by the HVAC units.

In one example, embodiment, a DCV-controlled HVAC system may include a gas sensor in a main return-air path installed nearest to its associated air-handling-unit (AHU) to detect concentration of one or more target gas type. However, this kind of arrangement may inherit a lagging response time of the AHU to react to demand of fresh air. For example, a sudden increase of number of people entering an indoor area may drastically deteriorate the quality of air of certain region in such area, however the gas sensor may only determine such change in the air quality until the concentration of different gases reach an equilibrium across different regions within the area, or when the pollutants reach the gas sensor after a relatively slow diffusion process.

This sluggishness may be more obvious in bigger a zone that an AHU is serving. In scenes like high-headroom spaces like auditorium hall or shopping mall atrium, where the return-air main dust may be mounted from few to a dozen meters high up above the occupancy, the response time can become tens of minutes. By the time the high-level-mounted gas sensor detected an occupancy in a zone, the latter might have be dismissed from that zone already, rendering change of fresh air rate meaningless, yielding a waste of energy spent and bad air experience in the occupancy.

With reference to Figure 1, there is shown a method 100 for managing indoor air quality in an area, e.g. a shopping mall or a commercial building including multiple zones, employing a HVAC unit (not shown) . The method 100 generally involves determining occupancy of an indoor area, thus determining a ventilation demand in the indoor area and controlling the HVAC unit. In step 102, occupancy of the indoor area is first determined. Preferably, this is carried out by determining the number of people in the indoor area.

In this embodiment, one or more images of the indoor area including individuals in the indoor area are captured. Then, the individuals in the images are identified so as to determine the number of people in the indoor area. It is preferred to identify the individuals and determine the number of people based on visual recognition of the individuals, such as face-detection and skeleton recognition, as shown in Figure 2. It is also preferred that the visual recognition is substantially real-time. Additionally or alternatively, Internet Protocol (IP) cameras with visual recognition techniques are used to capture the images such that these images and recognition results can be provided and uploaded to a server through wireless connections for storage for later use, as described in the following paragraphs.

Figure 2 shows an image obtained using visual skeleton recognition, illustrating the skeleton structures of the individuals in the image, thereby facilitating people counting. The skeleton recognition technique employed herein is known in the art, including informative region selection, feature extraction and object modelling. The algorithm is able to manage the occlusions, the crowds and the scale changes that may occur in the indoor area, allowing an accurate result for people count to be obtained.

Referring back to Figure 1, in step 104, the indoor air quality (IAQ) is determined. In particular, the concentration of carbon dioxide in the indoor area is determined by a carbon dioxide sensor, since carbon dioxide is the most common indicator for air quality related to human occupancy. In other embodiments, the IAQ may be determined by other IAQ monitors, such as a particulate matter (PM) sensor, a temperature sensor, a humidity sensor, or any combinations thereof. Similar to the above, in a preferred embodiment, the carbon dioxide sensor or other IAQ monitors are capable of connecting wirelessly to a server and providing detection results to the server for later use.

Preferably, a multiple-gas type monitoring apparatus may be used for determining the quality of air. For example, a gas sensor or detector may determine an air quality based on parameters such as PM2.5, relative humidity, carbon dioxide level and concentration of other gases.

Additionally, or optionally, the gas monitors may be deployed not only indoor but also in outdoor, so to generate smarter fresh air control decisions. For example, it is possible that air in an outdoor environment may be more polluted than the air in indoor area, e.g. in China and India metro cities as well as many regions with road-side facilities.

In step 106, a ventilation demand in the indoor area is determined, the ventilation demand being associated with the determined occupancy in step 102 and/or the determined concentration of carbon dioxide in step 104. The higher the occupancy and concentration of carbon dioxide, the greater the ventilation demand to exchange fresh air or outside air with the air indoor. Additionally, the ventilation demand may also be associated with air exchanges between the indoor area and outdoor environment separated by doors in the indoor area. The air exchanges may be used to determine an effect of indoor air quality by the server.

In step 108, the HVAC unit in the area is controlled with an operation profile so as to satisfy the determined ventilation demand in step 106. In one embodiment, this includes controlling and adjusting the speed of induction fans of the HVAC unit based on the determined occupancy and the determined concentration of carbon dioxide. In other embodiments, this may also include adjusting the ventilation rate, adjusting the ratio of the supply and extract air flow, modulating backup zonal vent fans to introduce fresh air and filter outside air, and sending alerts to an operator.

Referring to Figures 3A and 3B, two operation profiles of the HVAC unit are illustrated. Figure 3A shows a first operation profile where the speed of induction fans is set to a medium to high (Hi/Mid) speed. The HVAC unit is operated with the first operation profile when the determined ventilation demand is high, as shown with a high occupancy in Figure 3A. On the other hand, Figure 3B shows a second operation where the speed of induction fans is set to a zero to low (Low/Off) speed. The HVAC unit is operated with the second operation profile when the determined ventilation demand is low, as shown with a low occupancy in Figure 3B. As discussed above, although not shown in Figures 3A and 3B, the determined ventilation demand may also be associated with the concentration of carbon dioxide. As a result, the speed of fans may also increase when the determined concentration of carbon dioxide is high, and vice versa.

Also shown in Figures 3A and 3B is a user interface 300 on an external device displaying various information regarding the operation profile. For example, the speed of induction fan, the current time, temperature and humidity of the particular indoor area may be displayed on the user interface 300, details of which will be discussed below.

In an alternative embodiment, the HVAC unit may have more than two operation profiles, for example, the HVAC unit may have a third operation profile which allows both the speed of fans and the ratio of the supply and extract air flow to be adjusted at the same time, and a fourth operation profile which activates a backup zonal vent fan based on the determined ventilation demand.

Referring back to Figure 1, in step 110, status related to the indoor air quality, in particular the determined occupancy and concentration of carbon dioxide in

steps

102 and 104, is retrieved from the server and reported on a reporting dashboard on an external device for user interaction. Again, the external device is preferably wirelessly connected to the server such that status and data can be retrieved and displayed to the user anytime and anywhere. The user may manipulate the settings of the operation profile of the HVAC unit on the external device, to achieve better indoor air quality according to his/her needs.

Figure 4 shows a reporting dashboard 400 according to one example embodiment. Various information regarding the indoor environment is shown on the reporting dashboard 400. For example, the speed of induction fan, the current time, temperature and humidity of the indoor area may be displayed on the reporting dashboard 400. The reporting dashboard 400 also includes a user interface which allows the user to control and customise the operation profiles or settings of the HVAC unit. For example, the user may increase the speed of induction fan thus increase the indoor temperature when the current temperature is too high.

After the step of controlling the HVAC unit in step 108, in addition to the reporting of status on the reporting dashboard 400, the determined visual data and air quality status may also be aggregated and analysed on the server in step 112 (Figure 1) . The data may be aggregated on-premises or on cloud deployment. The analysed data is then displayed on an external device for user interaction. The external device may be the same external device displaying information regarding the operation profile in Figures 3A and 3B.

Additionally, a heat map may be generated based on the analysed data. For example, with reference to Figures 5A and 5B, two types of heat map may be generated for a HVAC unit employed in a shopping centre, representing the people count in different areas within the shopping centre. Different information regarding the people count, for example the people count across years (Figure 5A) and the people count at the same time in different areas (Figure 5B) , may be obtained. Advantageously, the heat map generated may correlate the promotional activities happening in stores in the shopping centre. This allows the user to control the HVAC unit to avoid a full-load operation, and avoid fresh air to be sent to low occupancy zones, based on the heat map.

With reference to Figure 6, there is shown a system 600 for managing indoor air quality in an area according to one example embodiment. The system 600 may be suitable for managing indoor air quality in an area including multiple zones, where each of the zones requires an individually managed air ventilation suitable for a dynamic occupancy of the indoor area. Preferably, the system 600 is operated using the method 100 as described above. Therefore, the system600 will be discussed hereinafter using the method 100 as an example embodiment. The system 600 includes a detection module 602 arranged to determine an occupancy of an indoor area in step 102, a processing module (not shown) arranged to determine a ventilation demand in the indoor area associated with the determined occupancy, and a controller module (not shown) arranged to control an HVAC unit in step 108with an operation profile so as to satisfy the determined ventilation demand.

Preferably, the detection module 602 includes a visual-based module, such as an IP camera, face-detection camera, a skeleton recognition camera, or a combination thereof. Figure 6 shows a real-time operation of the skeleton recognition camera 602. The camera 602 shares the captured image to the server wirelessly, allowing the captured image to be retrieved and displayed on an external device 604, e.g. a monitor of a desktop computer, to the user in real time.

In a more preferred embodiment, the system 600 includes more than one detection modules 602. In this embodiment, the plurality of detection modules 602 includes a first camera for identifying individuals based on face-detection and a second camera for identifying individuals based on skeleton recognition, such that the number of people in the indoor area is counted using both parameters. The detection modules 602 are further arranged to provide face-detection and visual skeleton recognition results to the server through wireless connections.

In one embodiment, step 104 may be performed by a gas sensor, for example a carbon dioxide sensor to determine a concentration of carbon dioxide in each of the zones in the indoor area.

The controller module of the system 600 may include a variable speed drive (VSD) controller in each zone arranged to control the speed of induction fans of the HVAC unit based on the determined ventilation demand by the processing module, which is based on the occupancy and the detected indoor air quality in a particular zone. The VSD controller is arranged to drive fans in the HVAC units with different operation profiles, and convert a constant air volume (CAV) fan to a variable air volume (VAV) fan. This allows fans to provide fresh air and circulated air to each of the zones in the indoor area based on different air ventilation requirements.

The detection module may also include a door switch arranged to detect movement of doors in the indoor area, such that the server can determine an effect of IAQ due to air exchanges between the indoor area and outdoor environment separated by the doors, thereby causing the controller module to manipulate the HVAC unit based on the determined effect. For example, in a situation where the temperature indoor is much lower than the outdoor environment, when the doors are frequently opened and closed, air exchanges may be rapid leading to a low IAQ. In this situation, the controller may control the HVAC unit to increase the speed of the induction fans, so as to maintain overall indoor temperature, and to achieve a better IAQ.

Additionally or alternatively, the system 600 may also include a lighting switch arranged to control lights in the indoor area based on the determined occupancy in step 102. The lighting switch may be connected to the server wirelessly and retrieve the determined occupancy from the server. Controls of the lights in the indoor area may include dimming, turning on or off the lights. This is particularly useful as it avoids lighting to be sent to low occupancy areas, resulting in higher energy efficiency.

In one embodiment, the system 600 may include a remote management platform 606, which may include the reporting dashboard 400 of Figure 4. The remote management platform 606 allows a user to interact with the server, to remotely control each of the HVAC units in different zones in the indoor area through the server and a cloud-based platform, and to customise the operation profiles in each of the individual zones associated with the occupancy of the indoor area, as discussed above.

These embodiments may be advantageous in that, the present invention provides a method to trigger HVAC events based on two parameters, people count and activity level, using an AI+IAQ approach on part-load changes with a fast response time (can be up to seconds) . Further, the present invention reduces operation expenditure (OPEX) while thermal comfort and statutory ventilation rate are maintained in the lower possible energy cost, offers low capacity expenditure (CAPEX) . The present invention also improves the accuracy of the conventional DCV method. In particular, the novel method eliminates false detection by using a skeleton recognition camera, as it can be mounted at high positions for large viewing coverage without omitting counting people wearing hats or masks in a distance.

It will be appreciated by persons skilled in the art that numerous variations and/or modifications may be made to the invention as shown in the specific embodiments without departing from the spirit or scope of the invention as broadly described. The present embodiments are, therefore, to be considered in all respects as illustrative and not restrictive.

Any reference to prior art contained herein is not to be taken as an admission that the information is common general knowledge, unless otherwise indicated.

Claims

A method for managing indoor air quality in an area, comprising the steps of:

- determining an occupancy of an indoor area;

- determining a ventilation demand in the indoor area associated with the determined occupancy; and

- controlling an HVAC unit with an operation profile so as to satisfy the determined ventilation demand.
The method according to claim 1, wherein the step of determining the occupancy includes determining number of people in the indoor area.
The method according to claim 2, wherein the step of determining number of people comprising the steps of:

- capturing one or more images of the indoor area including individuals in the indoor area; and

- identifying the individuals in the images so as to determine the number of people in the indoor area.
The method according to claim 3, wherein the determination of the number of people is based on visual skeleton recognition of the individuals in the images.
The method according to claim 1, further comprising the steps of:

- determining a concentration of carbon dioxide in the indoor area; and

- controlling the HVAC unit based on the determined concentration of carbon dioxide.
A system for managing indoor air quality in an area, comprising:

- a detection module arranged to determine an occupancy of an indoor area;

- a processing module arranged to determine a ventilation demand in the indoor area associated with the determined occupancy; and

- a controller module arranged to control an HVAC unit with an operation profile so as to satisfy the determined ventilation demand.
The system according to claim 6, wherein the detection module includes a face-detection camera.
The system according to claim 6, wherein the detection module includes a skeleton recognition camera.
The system according to claim 6, further comprising a gas sensor arranged to determine a concentration of carbon dioxide in the indoor area, wherein the controller module is further arranged to control the HVAC unit based on the determined concentration of carbon dioxide.
A system for managing indoor air quality in an area including multiple zones, each of the zones requires an individually managed air ventilation suitable for a dynamic occupancy of the indoor area, the system comprising:

- a plurality of visual-based people counting modules arranged to count a number of people in each of the zones in the indoor area, each of the visual-based people counting modules includes a first camera for identifying individuals based on face-detection and a second camera for identifying individuals based on skeleton recognition, the first and the second cameras are IP cameras that connect to a server through wireless connections;

- a plurality of indoor air quality sensors installed in each of the zones in the indoor area arranged to detect the air quality including a concentration of carbon dioxide in each of the zones in the indoor area, and to provide the detection results to the server;

- a plurality of variable speed drive (VSD) controllers arranged to drive fans in a plurality of HVAC units with different operation profiles for providing fresh air and circulated air to each of the zones in the indoor area based on different air ventilation requirements, wherein the VSD controllers in each zone control the speed of the fans based on the occupancy determined based on the number of people counted by the visual-based people counting module and the detected indoor air quality in a particular zone;

- a plurality of door switches arranged to detect doors in the indoor area being opened and closed, the server then determines an effect of indoor air quality due to air exchanges between the indoor area and outdoor environment separated by the doors; and

- a remote management platform that allows a user to interact with the server that performs data aggregation and analytics of visual data obtained from the visual-based people counting module and air quality status determined by the indoor air quality sensors; the remote management platform also allows the user to remotely control each of the HVAC units in different zones in the indoor area through the server and a cloud-based platform, and to customise the operation profiles in each of the individual zones associated with the occupancy of the indoor area.