WO2023175755A1 - Air conditioning system - Google Patents

Air conditioning system Download PDF

Info

Publication number
WO2023175755A1
WO2023175755A1 PCT/JP2022/011807 JP2022011807W WO2023175755A1 WO 2023175755 A1 WO2023175755 A1 WO 2023175755A1 JP 2022011807 W JP2022011807 W JP 2022011807W WO 2023175755 A1 WO2023175755 A1 WO 2023175755A1
Authority
WO
WIPO (PCT)
Prior art keywords
air conditioning
sensor
target area
concentration information
information
Prior art date
Application number
PCT/JP2022/011807
Other languages
French (fr)
Japanese (ja)
Inventor
敏基 吉田
Original Assignee
三菱電機株式会社
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 三菱電機株式会社 filed Critical 三菱電機株式会社
Priority to PCT/JP2022/011807 priority Critical patent/WO2023175755A1/en
Priority to JP2024507281A priority patent/JPWO2023175755A1/ja
Publication of WO2023175755A1 publication Critical patent/WO2023175755A1/en

Links

Images

Classifications

    • F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
    • F24—HEATING; RANGES; VENTILATING
    • F24F—AIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
    • F24F11/00—Control or safety arrangements
    • F24F11/62—Control or safety arrangements characterised by the type of control or by internal processing, e.g. using fuzzy logic, adaptive control or estimation of values
    • F24F11/63—Electronic processing
    • F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
    • F25—REFRIGERATION OR COOLING; COMBINED HEATING AND REFRIGERATION SYSTEMS; HEAT PUMP SYSTEMS; MANUFACTURE OR STORAGE OF ICE; LIQUEFACTION SOLIDIFICATION OF GASES
    • F25B—REFRIGERATION MACHINES, PLANTS OR SYSTEMS; COMBINED HEATING AND REFRIGERATION SYSTEMS; HEAT PUMP SYSTEMS
    • F25B49/00—Arrangement or mounting of control or safety devices
    • F25B49/02—Arrangement or mounting of control or safety devices for compression type machines, plants or systems

Definitions

  • the present disclosure relates to an air conditioning system that air-conditions a target area.
  • the present disclosure has been made in view of the above circumstances, and aims to provide an air conditioning system that can perform ventilation control that matches the characteristics of each target area.
  • An air conditioning system includes an air conditioning device that is indoor air conditioning equipment, an air conditioning controller that controls the air conditioning device, a CO2 sensor that is provided in each of a plurality of target areas and that acquires CO2 concentration information, and the CO2 and a BLE beacon that transmits position information indicating the position of the sensor, and the air conditioning controller includes a learning model that infers the future CO2 concentration information for each target area from the CO2 concentration information acquired by the CO2 sensor. If the future CO2 concentration information of the target area inferred by the learning model exceeds a threshold, the target area corresponds to the position indicated by the position information of the CO2 sensor transmitted from the BLE beacon.
  • the air conditioner is operated to perform air conditioning.
  • the target area corresponds to the position indicated by the position information of the CO2 sensor transmitted from the BLE beacon.
  • the air conditioner is operated to perform air conditioning. Thereby, it is possible to maintain an appropriate ventilation state for each area so that the CO2 concentration state in the target area is less than the reference value.
  • FIG. 1 is a diagram showing the configuration of an air conditioning system in Embodiment 1.
  • FIG. 1 is a diagram showing functional blocks of an air conditioning system according to Embodiment 1.
  • FIG. 3 is a diagram illustrating control of the air conditioning controller according to the first embodiment.
  • FIG. 3 is a diagram illustrating compulsory operation time calculation in the data processing unit according to the first embodiment. This is a learning and inference model used in both inference processing and learning processing in the data processing unit according to the first embodiment.
  • 3 is a flowchart of acquiring CO2 concentration information in the air conditioning system according to the first embodiment.
  • 3 is a flowchart of accumulation of CO2 concentration information of the air conditioning controller according to the first embodiment.
  • 2 is a flowchart of CO2 concentration excess detection processing of the air conditioning controller according to the first embodiment.
  • 7 is a flowchart of inference processing of the air conditioning controller according to the first embodiment.
  • 5 is a flowchart of learning processing of the air conditioning controller according to the first embodiment.
  • FIG. 1 is a diagram showing the configuration of an air conditioning system 100 in the first embodiment.
  • an air conditioning system 100 includes an outdoor unit 110, a ventilation device 120, an indoor unit 130, an IoT gateway 140, a human sensor 150, an air conditioning controller 160, an external memory 170, a cloud 180, a CO2 sensor 190, and a BLE (Bluetooth) Trademark) Low Energy) beacon 200.
  • the outdoor unit 110 is installed outdoors and connected to a ventilation system 120 and an indoor unit 130.
  • the outdoor unit 110 is connected to the indoor unit 130 through refrigerant piping, and supplies the indoor unit 130 with a refrigerant that has undergone heat exchange with the outside air.
  • the outdoor unit 110 is connected to the ventilation device 120 through air piping, takes in outside air and supplies it to the ventilation device 120, and exhausts indoor air sent from the ventilation device 120 to the outdoors.
  • the ventilation system 120 and the indoor unit 130 are air conditioners that are indoor air conditioning equipment.
  • the ventilation device 120 performs ventilation by exhausting indoor air to the outdoors and introducing outdoor air indoors.
  • the indoor unit 130 generates conditioned air whose temperature or humidity is adjusted by exchanging heat between the refrigerant sent from the outdoor unit 110 and indoor air, and supplies this conditioned air indoors.
  • the operation of the ventilation system 120 and the indoor unit 130 is controlled by an air conditioning controller 160.
  • An IoT gateway 140 is provided in the ventilation system 120 and exchanges data with the air conditioning controller 160. Furthermore, the IoT gateway 140 is provided in the indoor unit 130 and exchanges data with the air conditioning controller 160.
  • the human sensor 150 is provided in the ventilation system 120 and detects human information indicating the number of people and their positions in the target area. Further, the human sensor 150 is provided in the indoor unit 130 and detects human information indicating the number of people and positions in the target area. As this human sensor 150, for example, an image type human sensor is used. The increase in CO2 concentration is caused by the number and location of people, and this information is used in AI control described later.
  • the air conditioning controller 160 is a device for monitoring and controlling the plurality of ventilation devices 120, the outdoor unit 110, and the indoor unit 130. Additionally, the air conditioning controller 160 transmits and receives data to and from the IoT gateway 140 installed in the indoor unit 130 and the ventilation device 120.
  • the external memory 170 is a storage medium that stores data of the air conditioning controller 160.
  • the external memory 170 referred to here is a medium such as a USB memory or an SD card. This medium may be exchangeable by the user for one with a larger data capacity.
  • the cloud 180 is a data storage area that can be accessed over the Internet. This storage area is utilized to hold various data of the air conditioning controller 160.
  • the CO2 sensor 190 transmits CO2 concentration information in the target area via wireless communication.
  • the wireless communication here is assumed to be Wi-Fi, but wireless communication using BLE communication, ZigBee (registered trademark), and EnOcean may also be used.
  • the BLE beacon 200 transmits position information of the CO2 sensor 190 via BLE communication.
  • the BLE beacon 200 is used to grasp the position of the CO2 sensor 190.
  • the BLE beacon 200 is attached with a UUID (Universally Unique Identifier) and a unique identifier.
  • UUID Universally Unique Identifier
  • FIG. 2 is a diagram showing functional blocks of the air conditioning system 100 according to the first embodiment.
  • the IoT gateway 140 includes a BLE sensor information acquisition section 300, a CO2 data measurement section 310, and a data transmission/reception section 320. Although the IoT gateway 140 is described here, the function of the IoT gateway 140 may be added to the indoor unit 130 or the ventilation device 120.
  • the BLE sensor information acquisition unit 300 receives position information from the BLE beacon 200. At this time, the BLE sensor information acquisition unit 300 also receives the UUID and unique identifier attached to the BLE beacon 200.
  • the CO2 data measurement unit 310 acquires CO2 concentration information of the target area transmitted from the CO2 sensor 190.
  • the data transmission/reception unit 320 transmits the UUID and position information received by the BLE sensor information acquisition unit 300 and the CO2 concentration information of the target area acquired by the CO2 data measurement unit 310 to the air conditioning controller 160.
  • the air conditioning controller 160 periodically acquires the UUID, location information, and CO2 concentration information from the IoT gateway 140.
  • the air conditioning controller 160 includes a CO2 information acquisition section 210, a data processing section 220, a data non-volatile section 230, and an internal non-volatile area 240.
  • the CO2 information acquisition unit 210 acquires data from the IoT gateway 140. Data from the BLE sensor information acquisition unit 300 and CO2 data measurement unit 310 installed in the IoT gateway 140 is acquired via the data transmission/reception unit 320.
  • the data processing unit 220 performs data processing based on the information stored in the data non-volatile unit 230.
  • the data processing includes a process of making a judgment by comparing the obtained CO2 concentration and a CO2 concentration information threshold, and a process of calculating forced time using AI control (inference and learning).
  • the data processing unit 220 determines whether the CO2 concentration information exceeds a reference value.
  • the reference value is, for example, 1000 ppm. This reference value can be set in advance in the system.
  • the setting may be a common value for the entire system, or the reference value of the CO2 concentration information may be set individually for each CO2 sensor 190.
  • the ventilation device 120 is forced to operate in the target area of the CO2 sensor 190.
  • the forced operation refers to turning on the ventilation device 120 if it is in the OFF state, and refers to increasing the air volume if the ventilation device 120 is in the ON state.
  • the data non-volatile section 230 stores the data processed by the data processing section 220 in an internal non-volatile area 240 in a non-volatile manner.
  • the internal non-volatile area 240 stores CO2 concentration information and person information indicating the number of people and positions for each target area.
  • the nonvolatile storage destination for data may be the external nonvolatile area 250 or the cloud nonvolatile area 260.
  • the external non-volatile area 250 corresponds to external media such as an SD card and a USB memory.
  • the cloud non-volatile area 260 may be a drive on a network such as a LAN disk, in addition to a cloud service such as AWS (Amazon Web Services).
  • FIG. 3 is a diagram illustrating control of the air conditioning controller 160 according to the first embodiment.
  • the air conditioning controller 160 determines that forced operation of the ventilation system 120 and the indoor unit 130 is necessary, the ventilation system 120 and the indoor unit 130 adjacent to the target area where the CO2 concentration information has increased are , issue an operation command to ventilate the target area.
  • Figure 3 shows an example in which "strong wind” operation is performed only in the direction of the target area, which is the area with increased CO2 concentration, in order to ventilate only the area with increased CO2 concentration. ” control may also be used.
  • FIG. 4 is a diagram illustrating compulsory operation time calculation in the data processing unit 220 according to the first embodiment.
  • the CO2 concentration information of the target area, the CO2 concentration information threshold, and the number of people in the room are input as input parameters. Then, inference is made based on these input parameters to calculate the forced operation time.
  • the input parameter may include person information indicating the location of a person, which is room occupancy information in the target area.
  • the CO2 concentration information threshold value may be the same value as the above-mentioned reference value used for determining whether to forcefully operate the ventilation device 120 and the indoor unit 130, or may be a different value.
  • the CO2 concentration information threshold is set by the data processing unit 220.
  • the learning model md is used in the inference process.
  • the learning model md is provided for each target area, and infers future CO2 concentration information of the target area from the CO2 concentration information measured by the CO2 sensor 190. Specifically, inference processing is performed using accumulated data of past CO2 concentration information measured by the CO2 sensor 190, the number of people in the room, the CO2 concentration information threshold, and the learning model md, and , calculates the time until the CO2 concentration information reaches the threshold value. Then, based on the arrival time until the CO2 concentration reaches the threshold value, the time at which one or both of the ventilation device 120 and the indoor unit 130 will be forced to operate is calculated in order to prevent the CO2 concentration from exceeding the threshold value. The calculated time is output as the forced operation time.
  • Learning of the learning model md for each target area starts at a fixed time every day and is performed from past accumulated data of input data and arrival times.
  • FIG. 5 shows a learning and inference model used in both inference processing and learning processing in the data processing unit 220 according to the first embodiment.
  • the learning and inference model consists of an input layer I to which input parameters are input, an output layer O to which output parameters are output, a middle layer M_1, a middle layer M_2, and a middle layer M_3. Network.
  • the arrival time is calculated at the output layer O.
  • parameters, weights, and bias values in the intermediate layer M are updated from past accumulated data of input data and arrival times.
  • FIG. 6 is a flowchart of acquiring CO2 concentration information in the air conditioning system 100 according to the first embodiment.
  • the UUID and location information are periodically acquired from the BLE beacon 200 (step S101).
  • the UUID is an identifier of the BLE beacon 200 and is used for linking with the CO2 sensor 190.
  • CO2 concentration information at the location of the BLE beacon 200 is acquired by the CO2 sensor 190. Furthermore, the human sensor 150 acquires human information including the number of people and their positions (step S102).
  • the CO2 concentration information and person information are transmitted to the air conditioning controller 160 (step S103).
  • FIG. 7 is a flowchart of accumulation of CO2 concentration information in the air conditioning controller 160 according to the first embodiment.
  • the air conditioning controller 160 acquires BLE position information, UUID, CO2 concentration information, and person information indicating the number of people and their positions (step S201). Note that the acquisition is performed at regular intervals.
  • the air conditioning controller 160 compares the acquired UUID with a database that associates pre-stored UUIDs with the CO2 sensor 190, and performs individual identification of the CO2 sensor 190 (step S202).
  • One CO2 sensor 190 can be specified by the UUID.
  • the air conditioning controller 160 stores the obtained CO2 concentration information along with the time for each identified CO2 sensor 190 (step S203).
  • the air conditioning controller 160 determines whether the CO2 concentration information received in step S201 exceeds a reference value (step S204).
  • step S204 If it is determined in step S204 that the CO2 concentration information exceeds the reference value (YES in step S204), the process moves to the CO2 concentration excess process in FIG. 8 (step S205).
  • step S204 if it is determined in step S204 that the received CO2 concentration information does not exceed the reference value (NO in step S204), the CO2 concentration information accumulation process of FIG. 7 is ended.
  • FIG. 8 is a flowchart of the CO2 concentration excess processing of the air conditioning controller 160 according to the first embodiment.
  • the air conditioning controller 160 identifies the ventilation device 120 and indoor unit 130 near the target area using the BLE position information (step S301).
  • the air conditioning controller 160 determines whether the ventilation device 120 and the indoor unit 130 near the target area are in the operating state (step S302).
  • step S302 if it is determined that the operating state of the ventilation device 120 and the indoor unit 130 is not in the operating state (NO in step S302), the operation of the ventilation device 120 and the indoor unit 130 is turned on and forced ventilation operation is performed ( Step S303).
  • step S302 if it is determined that the operating state of the ventilation device 120 and the indoor unit 130 is in the operating state (YES in step S302), the air volume of the ventilation device 120 and the indoor unit 130 in the target area direction is increased (step S304).
  • FIG. 9 is a flowchart of the inference process of the air conditioning controller 160 according to the first embodiment.
  • the air conditioning controller 160 determines whether or not it is necessary to start inference (step S401). Specifically, the air conditioning controller 160 determines that it is necessary to start inference when the CO2 concentration information exceeds an internal lower limit of 500 ppm, for example. This is because in a low concentration state, there is no need for ventilation, so a certain lower limit value is set and the inference process is started at the timing when the CO2 concentration information has increased to a certain extent.
  • the internal lower limit value is a value lower than the reference value and the CO2 concentration information threshold value. As input parameters, CO2 concentration information, CO2 concentration information threshold, and the number of people in the room are input.
  • step S401 If it is determined in step S401 that it is not necessary to start inferring CO2 concentration information (NO in step S401), the process in FIG. 9 ends.
  • step S401 If it is determined in step S401 that it is necessary to start inference of CO2 concentration information (YES in step S401), inference processing of CO2 concentration information is performed (step S402). In step S402, inference processing from the input layer I to the output layer O shown in FIG. 5 is performed. Then, the time until the CO2 concentration information exceeds the CO2 concentration information threshold, that is, the arrival time shown in FIG. 4 is calculated.
  • step S403 it is determined whether the inference result is expected to exceed the CO2 concentration information threshold.
  • the arrival time obtained in step S402 is shorter than the time until the next step S401 is executed, it is determined that the scheduled time has been exceeded. That is, it is determined whether the CO2 concentration exceeds the CO2 concentration information threshold earlier than the next timing of the periodically performed inference process.
  • step S403 if the inference result is not expected to exceed the CO2 concentration information threshold (NO in step S403), the process of the inference process flowchart in FIG. 9 ends.
  • step S403 if the inference result is scheduled to exceed the CO2 concentration information threshold (YES in step S403), at the forced operation time explained in FIG. (Step S404), the process ends. Specifically, in step S404, if the CO2 concentration information of the target area inferred by the learning model md exceeds the CO2 concentration information threshold, the air conditioning controller 160 uses the position information of the CO2 sensor 190 transmitted from the BLE beacon. Ventilate the air conditioner in the target area corresponding to the location indicated by. Note that when the ventilation device 120 and the indoor unit 130 are already in operation, the air volume of the ventilation device 120 and the indoor unit 130 is increased.
  • FIG. 10 is a flowchart of the learning process of the air conditioning controller 160 according to the first embodiment.
  • the air conditioning controller 160 periodically determines whether it is learning start time (step S501). Since learning increases the CPU load, the air conditioning controller 160 starts learning at a predetermined time once or twice a day.
  • the air conditioning controller 160 performs the learning process from the output layer O to the input layer I described in FIG. 5 for each area (step S502).
  • the air conditioning controller 160 updates the parameters, weights, and bias values in the intermediate layer M_1, intermediate layer M_2, and intermediate layer M_3 according to the learning results (step S503).
  • the air conditioning controller 160 performs learning of the learning model md based on the person information and CO2 concentration information stored in the internal nonvolatile area 240, which is a storage unit.
  • the air conditioning system 100 of the first embodiment when the CO2 concentration information inferred in the target area exceeds the CO2 concentration information threshold using the learning model md, the CO2 sensor 190 transmitted from the BLE beacon The air conditioner is operated to perform air conditioning in the target area corresponding to the position indicated by the position information. Thereby, it is possible to maintain an appropriate air conditioning state for each area so that the CO2 concentration state in the target area is less than the reference value.
  • Air conditioning system 110 Outdoor unit, 120 Ventilation system, 130 Indoor unit, 140 IoT gateway, 150 Human sensor, 160 Air conditioning controller, 170 External memory, 180 Cloud, 190 CO2 sensor, 200 BLE beacon, 210 CO2 information acquisition unit , 220 data processing unit, 230 data nonvolatile unit, 240 internal nonvolatile area, 250 external nonvolatile area, 260 cloud nonvolatile area, 300 BLE sensor information acquisition unit, 310 CO2 data measurement unit, 320 data transmission/reception unit, I input phase, O output layer , M_1, M_2, M_3 middle layer, md learning model.

Landscapes

  • Engineering & Computer Science (AREA)
  • Signal Processing (AREA)
  • Physics & Mathematics (AREA)
  • Mechanical Engineering (AREA)
  • General Engineering & Computer Science (AREA)
  • Fuzzy Systems (AREA)
  • Mathematical Physics (AREA)
  • Chemical & Material Sciences (AREA)
  • Combustion & Propulsion (AREA)
  • Thermal Sciences (AREA)
  • Air Conditioning Control Device (AREA)

Abstract

This air conditioning system comprises: an air conditioner that is indoor air conditioning equipment; an air conditioning controller that controls the air conditioner; a CO2 sensor provided in each of a plurality of target areas and acquiring CO2 concentration information; and a BLE beacon that transmits location information indicating the location of the CO2 sensor. The air conditioning controller has a learning model that infers future CO2 concentration information for each target area from the CO2 concentration information acquired by the CO2 sensor, and when the future CO2 concentration information of the target area inferred by the learning model exceeds a threshold, the air conditioning controller operates the air conditioner to perform air-conditioning of the target area corresponding to the location indicated by the location information of the CO2 sensor transmitted from the BLE beacon.

Description

空調システムair conditioning system
　本開示は、対象エリアの空調を行う空調システムに関する。 The present disclosure relates to an air conditioning system that air-conditions a target area.
　従来、汚染物質濃度が第１濃度以上第２濃度未満であれば、空調機を制御するという技術が知られている（特許文献１参照）。 Conventionally, a technique is known in which an air conditioner is controlled when the pollutant concentration is at least a first concentration and less than a second concentration (see Patent Document 1).
国際公開第２０２０／２５５８７５号International Publication No. 2020/255875
　しかし、汚染物質濃度の測定は、その時点の値でのみ判断しているものであり、エリア毎の特性に見合った換気制御でない。このような手法は、「基準値を超えたら」というその条件発生後の事後的な対応となっていて、その基準値を超えないように屋内を換気するという手法ではない。 However, measurement of pollutant concentration is based only on the value at that point in time, and ventilation control is not appropriate to the characteristics of each area. Such a method is a reactive response after the condition occurs "if the standard value is exceeded," and is not a method of ventilating the room to prevent the standard value from being exceeded.
　本開示は、上記実情に鑑みてなされたものであり、対象エリア毎の特性に合った換気制御を行うことができる空調システムを提供することを目的とする。 The present disclosure has been made in view of the above circumstances, and aims to provide an air conditioning system that can perform ventilation control that matches the characteristics of each target area.
　本開示に係る空調システムは、屋内の空調設備である空調装置と、前記空調装置を制御する空調コントローラと、複数の対象エリアのそれぞれに設けられ、ＣＯ２濃度情報を取得するＣＯ２センサと、前記ＣＯ２センサの位置を示す位置情報を発信するＢＬＥビーコンとを具備し、前記空調コントローラは、前記ＣＯ２センサにより取得された前記ＣＯ２濃度情報から前記対象エリア毎の将来の前記ＣＯ２濃度情報を推論する学習モデルを具備し、前記学習モデルにより推論された前記対象エリアの前記将来のＣＯ２濃度情報が閾値を超える場合、前記ＢＬＥビーコンから発信された前記ＣＯ２センサの位置情報によって示される位置に対応する前記対象エリアの空気調和を行うように前記空調装置を動作させる。 An air conditioning system according to the present disclosure includes an air conditioning device that is indoor air conditioning equipment, an air conditioning controller that controls the air conditioning device, a CO2 sensor that is provided in each of a plurality of target areas and that acquires CO2 concentration information, and the CO2 and a BLE beacon that transmits position information indicating the position of the sensor, and the air conditioning controller includes a learning model that infers the future CO2 concentration information for each target area from the CO2 concentration information acquired by the CO2 sensor. If the future CO2 concentration information of the target area inferred by the learning model exceeds a threshold, the target area corresponds to the position indicated by the position information of the CO2 sensor transmitted from the BLE beacon. The air conditioner is operated to perform air conditioning.
　本開示によれば、学習モデルを使用して、推論された対象エリアのＣＯ２濃度情報が閾値を超えている場合、ＢＬＥビーコンから発信されたＣＯ２センサの位置情報によって示される位置に対応する対象エリアの空気調和を行うように空調装置を動作させる。これにより、対象エリアのＣＯ２濃度状態が基準値未満となるように、エリア毎に沿った適切な換気状態を維持することができる。 According to the present disclosure, if the CO2 concentration information of the target area inferred using the learning model exceeds the threshold, the target area corresponds to the position indicated by the position information of the CO2 sensor transmitted from the BLE beacon. The air conditioner is operated to perform air conditioning. Thereby, it is possible to maintain an appropriate ventilation state for each area so that the CO2 concentration state in the target area is less than the reference value.
実施形態１における空調システムの構成を示す図である。1 is a diagram showing the configuration of an air conditioning system in Embodiment 1. FIG. 実施形態１に係る空調システムの機能ブロックを示す図である。1 is a diagram showing functional blocks of an air conditioning system according to Embodiment 1. FIG. 実施形態１に係る空調コントローラの制御を説明する図である。FIG. 3 is a diagram illustrating control of the air conditioning controller according to the first embodiment. 実施形態１に係るデータ処理部における強制動作時刻算出を説明する図である。FIG. 3 is a diagram illustrating compulsory operation time calculation in the data processing unit according to the first embodiment. 実施形態１に係るデータ処理部における推論処理と学習処理の両方で使用する学習及び推論モデルである。This is a learning and inference model used in both inference processing and learning processing in the data processing unit according to the first embodiment. 実施形態１に係る空調システムでのＣＯ２濃度情報取得のフローチャートである。3 is a flowchart of acquiring CO2 concentration information in the air conditioning system according to the first embodiment. 実施形態１に係る空調コントローラのＣＯ２濃度情報の蓄積のフローチャートである。3 is a flowchart of accumulation of CO2 concentration information of the air conditioning controller according to the first embodiment. 実施形態１に係る空調コントローラのＣＯ２濃度超過検知処理のフローチャートである。2 is a flowchart of CO2 concentration excess detection processing of the air conditioning controller according to the first embodiment. 実施形態１に係る空調コントローラの推論処理のフローチャートである。7 is a flowchart of inference processing of the air conditioning controller according to the first embodiment. 実施形態１に係る空調コントローラの学習処理のフローチャートである。5 is a flowchart of learning processing of the air conditioning controller according to the first embodiment.
　以下、図面を参照して、実施形態に係る空調システム１００について説明する。なお、図面において、同一の構成要素には同一符号を付して説明し、重複説明は必要な場合にのみ行う。 Hereinafter, an air conditioning system 100 according to an embodiment will be described with reference to the drawings. In addition, in the drawings, the same components will be described with the same reference numerals, and repeated description will be given only when necessary.
実施形態１．
　図１は、実施形態１における空調システム１００の構成を示す図である。図１において、空調システム１００は、室外機１１０、換気装置１２０、室内機１３０、ＩｏＴゲートウェイ１４０、人感センサ１５０、空調コントローラ１６０、外部メモリ１７０、クラウド１８０、ＣＯ２センサ１９０及びＢＬＥ（Bluetooth(登録商標） Low Energy）ビーコン２００を備える。 Embodiment 1.
FIG. 1 is a diagram showing the configuration of an air conditioning system 100 in the first embodiment. In FIG. 1, an air conditioning system 100 includes an outdoor unit 110, a ventilation device 120, an indoor unit 130, an IoT gateway 140, a human sensor 150, an air conditioning controller 160, an external memory 170, a cloud 180, a CO2 sensor 190, and a BLE (Bluetooth) Trademark) Low Energy) beacon 200.
　室外機１１０は、屋外に設置され、換気装置１２０と、室内機１３０とに接続されている。室外機１１０は、室内機１３０と冷媒配管で接続されており、外気と熱交換した冷媒を室内機１３０に供給する。室外機１１０は、換気装置１２０と空気配管で接続されており、外気を吸気して換気装置１２０に供給するとともに、換気装置１２０から送られた屋内の空気を屋外に排気する。 The outdoor unit 110 is installed outdoors and connected to a ventilation system 120 and an indoor unit 130. The outdoor unit 110 is connected to the indoor unit 130 through refrigerant piping, and supplies the indoor unit 130 with a refrigerant that has undergone heat exchange with the outside air. The outdoor unit 110 is connected to the ventilation device 120 through air piping, takes in outside air and supplies it to the ventilation device 120, and exhausts indoor air sent from the ventilation device 120 to the outdoors.
　換気装置１２０及び室内機１３０は、屋内の空調設備である空調装置である。換気装置１２０は、屋内の空気を屋外に排出するとともに、屋外の空気を屋内に導入して換気を行う。室内機１３０は、室外機１１０から送られた冷媒と屋内の空気とを熱交換させることで温度又は湿度が調節された調和空気を生成し、この調和空気を屋内に供給する。換気装置１２０及び室内機１３０は、空調コントローラ１６０によって動作制御される。 The ventilation system 120 and the indoor unit 130 are air conditioners that are indoor air conditioning equipment. The ventilation device 120 performs ventilation by exhausting indoor air to the outdoors and introducing outdoor air indoors. The indoor unit 130 generates conditioned air whose temperature or humidity is adjusted by exchanging heat between the refrigerant sent from the outdoor unit 110 and indoor air, and supplies this conditioned air indoors. The operation of the ventilation system 120 and the indoor unit 130 is controlled by an air conditioning controller 160.
　ＩｏＴ（Internet of things）ゲートウェイ１４０は、換気装置１２０に備えられ、空調コントローラ１６０とデータの受け渡しをする。また、ＩｏＴゲートウェイ１４０は、室内機１３０に備えられ、空調コントローラ１６０とデータの受け渡しをする。 An IoT (Internet of things) gateway 140 is provided in the ventilation system 120 and exchanges data with the air conditioning controller 160. Furthermore, the IoT gateway 140 is provided in the indoor unit 130 and exchanges data with the air conditioning controller 160.
　人感センサ１５０は、換気装置１２０に備えられ、対象エリアの人数と位置と示す人情報を検出する。また、人感センサ１５０は、室内機１３０に備えられ、対象エリアの人数と位置と示す人情報を検出する。この人感センサ１５０として、例えば、画像型人感センサが用いられる。ＣＯ２濃度の上昇要因は、人の人数と位置とに起因するものであり、この情報は、後述のＡＩ制御で用いる。 The human sensor 150 is provided in the ventilation system 120 and detects human information indicating the number of people and their positions in the target area. Further, the human sensor 150 is provided in the indoor unit 130 and detects human information indicating the number of people and positions in the target area. As this human sensor 150, for example, an image type human sensor is used. The increase in CO2 concentration is caused by the number and location of people, and this information is used in AI control described later.
　空調コントローラ１６０は、複数の換気装置１２０、室外機１１０及び室内機１３０の監視及び制御を行うための装置である。また、空調コントローラ１６０は、室内機１３０及び換気装置１２０に搭載されたＩｏＴゲートウェイ１４０との間でデータを送受信する。 The air conditioning controller 160 is a device for monitoring and controlling the plurality of ventilation devices 120, the outdoor unit 110, and the indoor unit 130. Additionally, the air conditioning controller 160 transmits and receives data to and from the IoT gateway 140 installed in the indoor unit 130 and the ventilation device 120.
　外部メモリ１７０は、空調コントローラ１６０のデータを蓄積する記憶媒体である。ここで言う外部メモリ１７０は、ＵＳＢメモリ及びＳＤカードといった媒体である。この媒体はユーザによりデータ容量のより大きいものへの交換を可能としてもよい。 The external memory 170 is a storage medium that stores data of the air conditioning controller 160. The external memory 170 referred to here is a medium such as a USB memory or an SD card. This medium may be exchangeable by the user for one with a larger data capacity.
　クラウド１８０は、インターネットにてアクセス可能なデータ保存領域である。この保存領域は、空調コントローラ１６０の各種データを保持するために活用される。 The cloud 180 is a data storage area that can be accessed over the Internet. This storage area is utilized to hold various data of the air conditioning controller 160.
　ＣＯ２センサ１９０は、対象エリアのＣＯ２濃度情報を無線通信にて発信するものである。ここでの無線通信は、Ｗｉ－Ｆｉを想定しているが、ＢＬＥ通信、ＺｉｇＢｅｅ（登録商標）及びＥｎＯｃｅａｎによる無線通信としてもよい。 The CO2 sensor 190 transmits CO2 concentration information in the target area via wireless communication. The wireless communication here is assumed to be Wi-Fi, but wireless communication using BLE communication, ZigBee (registered trademark), and EnOcean may also be used.
　ＢＬＥビーコン２００は、ＣＯ２センサ１９０の位置情報をＢＬＥ通信にて発信するものである。ここでは、ＢＬＥビーコン２００をＣＯ２センサ１９０と一体に配置することで、ＣＯ２センサ１９０の位置を把握するのに用いられる。ＢＬＥビーコン２００には、ＵＵＩＤ（Universally Unique Identifier）とユニークな識別子が付帯している。 The BLE beacon 200 transmits position information of the CO2 sensor 190 via BLE communication. Here, by disposing the BLE beacon 200 integrally with the CO2 sensor 190, it is used to grasp the position of the CO2 sensor 190. The BLE beacon 200 is attached with a UUID (Universally Unique Identifier) and a unique identifier.
　図２は、実施形態１に係る空調システム１００の機能ブロックを示す図である。 FIG. 2 is a diagram showing functional blocks of the air conditioning system 100 according to the first embodiment.
　ＩｏＴゲートウェイ１４０は、ＢＬＥセンサ情報取得部３００、ＣＯ２データ測定部３１０及びデータ送受信部３２０を有している。ここでは、ＩｏＴゲートウェイ１４０と記載しているが、室内機１３０又は換気装置１２０にＩｏＴゲートウェイ１４０の機能を付加してもよい。 The IoT gateway 140 includes a BLE sensor information acquisition section 300, a CO2 data measurement section 310, and a data transmission/reception section 320. Although the IoT gateway 140 is described here, the function of the IoT gateway 140 may be added to the indoor unit 130 or the ventilation device 120.
　ＢＬＥセンサ情報取得部３００は、ＢＬＥビーコン２００からの位置情報を受信する。この時、ＢＬＥセンサ情報取得部３００は、併せてＢＬＥビーコン２００に付帯されるＵＵＩＤとユニークな識別子も受信する。 The BLE sensor information acquisition unit 300 receives position information from the BLE beacon 200. At this time, the BLE sensor information acquisition unit 300 also receives the UUID and unique identifier attached to the BLE beacon 200.
　ＣＯ２データ測定部３１０は、ＣＯ２センサ１９０から発信された対象エリアのＣＯ２濃度情報を取得する。 The CO2 data measurement unit 310 acquires CO2 concentration information of the target area transmitted from the CO2 sensor 190.
　データ送受信部３２０は、ＢＬＥセンサ情報取得部３００で受信されたＵＵＩＤ及び位置情報、ＣＯ２データ測定部３１０で取得された対象エリアのＣＯ２濃度情報を空調コントローラ１６０に送信する。 The data transmission/reception unit 320 transmits the UUID and position information received by the BLE sensor information acquisition unit 300 and the CO2 concentration information of the target area acquired by the CO2 data measurement unit 310 to the air conditioning controller 160.
　空調コントローラ１６０は、ＩｏＴゲートウェイ１４０より定期的に、ＵＵＩＤ及び位置情報とＣＯ２濃度情報とを取得する。空調コントローラ１６０は、ＣＯ２情報取得部２１０、データ処理部２２０、データ不揮発部２３０及び内部不揮発領域２４０を有する。 The air conditioning controller 160 periodically acquires the UUID, location information, and CO2 concentration information from the IoT gateway 140. The air conditioning controller 160 includes a CO2 information acquisition section 210, a data processing section 220, a data non-volatile section 230, and an internal non-volatile area 240.
　ＣＯ２情報取得部２１０は、ＩｏＴゲートウェイ１４０からデータ取得を行う。ＩｏＴゲートウェイ１４０に搭載されているＢＬＥセンサ情報取得部３００及びＣＯ２データ測定部３１０からのデータを、データ送受信部３２０を介して取得する。 The CO2 information acquisition unit 210 acquires data from the IoT gateway 140. Data from the BLE sensor information acquisition unit 300 and CO2 data measurement unit 310 installed in the IoT gateway 140 is acquired via the data transmission/reception unit 320.
　データ処理部２２０は、データ不揮発部２３０に記憶された情報を元に、データ処理を行う。データ処理として、取得したＣＯ２濃度とＣＯ２濃度情報閾値とを比較して判断を行う処理と、ＡＩ制御（推論及び学習）により強制時時刻算出を行う処理とがある。 The data processing unit 220 performs data processing based on the information stored in the data non-volatile unit 230. The data processing includes a process of making a judgment by comparing the obtained CO2 concentration and a CO2 concentration information threshold, and a process of calculating forced time using AI control (inference and learning).
　データ処理部２２０は、データを保存する際、ＣＯ２濃度情報が基準値を超過していないか判断する。基準値は、例えば、１０００ｐｐｍである。この基準値については、システムで予め設定できるようにしておく。設定は、システム全体での共通値としてもよいし、ＣＯ２センサ１９０毎に個別にＣＯ２濃度情報の基準値を設定できるようにしてもよい。 When saving data, the data processing unit 220 determines whether the CO2 concentration information exceeds a reference value. The reference value is, for example, 1000 ppm. This reference value can be set in advance in the system. The setting may be a common value for the entire system, or the reference value of the CO2 concentration information may be set individually for each CO2 sensor 190.
　データ処理部２２０がＣＯ２濃度の基準値超過を検出した場合は、そのＣＯ２センサ１９０の対象エリアに対して、換気装置１２０を強制動作させる。強制動作とは、換気装置１２０がＯＦＦ状態であれば運転ＯＮとすることを指し、換気装置１２０が運転ＯＮ状態であれば風量の増加を指す。 If the data processing unit 220 detects that the CO2 concentration exceeds the reference value, the ventilation device 120 is forced to operate in the target area of the CO2 sensor 190. The forced operation refers to turning on the ventilation device 120 if it is in the OFF state, and refers to increasing the air volume if the ventilation device 120 is in the ON state.
　ＡＩ制御（推論及び学習）により強制時時刻算出を行う処理については、後に図４を参照して説明する。 The process of calculating forced time using AI control (inference and learning) will be described later with reference to FIG. 4.
　データ不揮発部２３０は、データ処理部２２０が処理したデータを、内部不揮発領域２４０へ不揮発保存する。内部不揮発領域２４０は、対象エリア毎にＣＯ２濃度情報、人数と位置とを示す人情報を蓄積する。 The data non-volatile section 230 stores the data processed by the data processing section 220 in an internal non-volatile area 240 in a non-volatile manner. The internal non-volatile area 240 stores CO2 concentration information and person information indicating the number of people and positions for each target area.
　データの不揮発保存先は、内部不揮発領域２４０以外にも、外部不揮発領域２５０としてもよいし、クラウド不揮発領域２６０としてもよい。外部不揮発領域２５０は、ＳＤカード及びＵＳＢメモリといった外部メディアが該当する。クラウド不揮発領域２６０は、ＡＷＳ（アマゾンウェブサービス）などのクラウドサービスの他に、ＬＡＮ　ＤＩＳＫのようなネットワーク上のドライブとしてもよい。 In addition to the internal nonvolatile area 240, the nonvolatile storage destination for data may be the external nonvolatile area 250 or the cloud nonvolatile area 260. The external non-volatile area 250 corresponds to external media such as an SD card and a USB memory. The cloud non-volatile area 260 may be a drive on a network such as a LAN disk, in addition to a cloud service such as AWS (Amazon Web Services).
　図３は、実施形態１に係る空調コントローラ１６０の制御を説明する図である。 FIG. 3 is a diagram illustrating control of the air conditioning controller 160 according to the first embodiment.
　図３に示すように、空調コントローラ１６０が、換気装置１２０及び室内機１３０の強制動作が必要と判断した場合、ＣＯ２濃度情報が上昇した対象エリアに隣接する換気装置１２０及び室内機１３０に対して、対象エリアの換気を行う動作命令を行う。 As shown in FIG. 3, when the air conditioning controller 160 determines that forced operation of the ventilation system 120 and the indoor unit 130 is necessary, the ventilation system 120 and the indoor unit 130 adjacent to the target area where the CO2 concentration information has increased are , issue an operation command to ventilate the target area.
　図３は、ＣＯ２濃度上昇エリアのみの換気目的のため、対象エリアであるＣＯ２濃度上昇エリアの方向のみを「強風」動作にした例であるが、部屋全体の換気のために、全方向「強風」とする制御としても良い。 Figure 3 shows an example in which "strong wind" operation is performed only in the direction of the target area, which is the area with increased CO2 concentration, in order to ventilate only the area with increased CO2 concentration. ” control may also be used.
　図４は、実施形態１に係るデータ処理部２２０における強制動作時刻算出を説明する図である。 FIG. 4 is a diagram illustrating compulsory operation time calculation in the data processing unit 220 according to the first embodiment.
　図４に示すように、強制動作時刻算出は、入力パラメータとして、対象エリアのＣＯ２濃度情報、ＣＯ２濃度情報閾値及び在室人数が入力される。そして、これら入力パラメータを元に推論を行い、強制動作時刻を算出する。なお、入力パラメータとして、対象エリアの在室情報である人の位置を示す人情報を含んでいてもよい。 As shown in FIG. 4, for forced operation time calculation, the CO2 concentration information of the target area, the CO2 concentration information threshold, and the number of people in the room are input as input parameters. Then, inference is made based on these input parameters to calculate the forced operation time. Note that the input parameter may include person information indicating the location of a person, which is room occupancy information in the target area.
　ＣＯ２濃度情報は、内部不揮発領域２４０に保存されている時系列のデータを用いる。ＣＯ２濃度情報閾値は、換気装置１２０及び室内機１３０を強制動作させるか否かの判断に使用される上述の基準値と同じ値であってもよいし、異なる値であってもよい。ＣＯ２濃度情報閾値は、データ処理部２２０で設定される。推論処理では、学習モデルｍｄを利用する。 For the CO2 concentration information, time-series data stored in the internal non-volatile area 240 is used. The CO2 concentration information threshold value may be the same value as the above-mentioned reference value used for determining whether to forcefully operate the ventilation device 120 and the indoor unit 130, or may be a different value. The CO2 concentration information threshold is set by the data processing unit 220. The learning model md is used in the inference process.
　学習モデルｍｄは、対象エリア毎に設けられ、ＣＯ２センサ１９０により測定されたＣＯ２濃度情報から対象エリアの将来のＣＯ２濃度情報を推論する。具体的には、ＣＯ２センサ１９０により測定された過去のＣＯ２濃度情報の蓄積データと、在室人数と、ＣＯ２濃度情報閾値と、学習モデルｍｄとを用いた推論処理を行って、当該対象エリアにおいて、今後ＣＯ２濃度情報が閾値に到達までの時間を算出する。そして、ＣＯ２濃度が閾値に到達するまでの到達時間から、ＣＯ２濃度が閾値を超えないようにするために換気装置１２０及び室内機１３０のいずれか又は両方を強制的に動作させる時刻を計算する。計算された時刻が、強制動作時刻として出力される。 The learning model md is provided for each target area, and infers future CO2 concentration information of the target area from the CO2 concentration information measured by the CO2 sensor 190. Specifically, inference processing is performed using accumulated data of past CO2 concentration information measured by the CO2 sensor 190, the number of people in the room, the CO2 concentration information threshold, and the learning model md, and , calculates the time until the CO2 concentration information reaches the threshold value. Then, based on the arrival time until the CO2 concentration reaches the threshold value, the time at which one or both of the ventilation device 120 and the indoor unit 130 will be forced to operate is calculated in order to prevent the CO2 concentration from exceeding the threshold value. The calculated time is output as the forced operation time.
　各対象エリアの学習モデルｍｄの学習は、毎日決まった時間に開始し、入力データと到達時間との過去の蓄積データから行う。 Learning of the learning model md for each target area starts at a fixed time every day and is performed from past accumulated data of input data and arrival times.
　図５は、実施形態１に係るデータ処理部２２０における推論処理と学習処理の両方で使用する学習及び推論モデルである。 FIG. 5 shows a learning and inference model used in both inference processing and learning processing in the data processing unit 220 according to the first embodiment.
　図５に示すように、学習及び推論モデルは、入力パラメータが入力される入力層Ｉ、出力パラメータが出力される出力層Ｏ、中間層Ｍ＿１、中間層Ｍ＿２及び中間層Ｍ＿３から構成されるＤｅｅｐ　Ｎｅｕｒａｌ　Ｎｅｔｗｏｒｋである。 As shown in FIG. 5, the learning and inference model consists of an input layer I to which input parameters are input, an output layer O to which output parameters are output, a middle layer M_1, a middle layer M_2, and a middle layer M_3. Network.
　推論処理では、入力層Ｉからスタートし、出力層Ｏで到達時間を算出する。学習処理では、入力データと到達時間との過去の蓄積データから、中間層Ｍでのパラメータ、重み及びバイアス値を更新する。 In the inference process, starting from the input layer I, the arrival time is calculated at the output layer O. In the learning process, parameters, weights, and bias values in the intermediate layer M are updated from past accumulated data of input data and arrival times.
　次に、空調システム１００の動作の詳細をフローチャートにて説明する。図６は、実施形態１に係る空調システム１００でのＣＯ２濃度情報取得のフローチャートである。 Next, details of the operation of the air conditioning system 100 will be explained using a flowchart. FIG. 6 is a flowchart of acquiring CO2 concentration information in the air conditioning system 100 according to the first embodiment.
　図６において、定期的にＢＬＥビーコン２００より、ＵＵＩＤ、及び位置情報を取得する（ステップＳ１０１）。ＵＵＩＤは、ＢＬＥビーコン２００の識別子であり、ＣＯ２センサ１９０との紐付けに利用する。 In FIG. 6, the UUID and location information are periodically acquired from the BLE beacon 200 (step S101). The UUID is an identifier of the BLE beacon 200 and is used for linking with the CO2 sensor 190.
　次に、ＢＬＥビーコン２００の場所のＣＯ２濃度情報をＣＯ２センサ１９０により、取得する。また、人感センサ１５０により、人数と位置との人情報を取得する（ステップＳ１０２）。 Next, CO2 concentration information at the location of the BLE beacon 200 is acquired by the CO2 sensor 190. Furthermore, the human sensor 150 acquires human information including the number of people and their positions (step S102).
　次に、ＣＯ２濃度情報及び人情報を空調コントローラ１６０へ送信する（ステップＳ１０３）。 Next, the CO2 concentration information and person information are transmitted to the air conditioning controller 160 (step S103).
　図７は、実施形態１に係る空調コントローラ１６０のＣＯ２濃度情報の蓄積のフローチャートである。 FIG. 7 is a flowchart of accumulation of CO2 concentration information in the air conditioning controller 160 according to the first embodiment.
　空調コントローラ１６０は、よりＢＬＥ位置情報、ＵＵＩＤ、ＣＯ２濃度情報及び人数と位置を示す人情報を取得する（ステップＳ２０１）。なお、取得は一定時間毎に実施する。 The air conditioning controller 160 acquires BLE position information, UUID, CO2 concentration information, and person information indicating the number of people and their positions (step S201). Note that the acquisition is performed at regular intervals.
　空調コントローラ１６０は、取得したＵＵＩＤを予め記憶されたＵＵＩＤとＣＯ２センサ１９０とを対応づけたデータベースと照合し、ＣＯ２センサ１９０の個体識別を行う（ステップＳ２０２）。ＵＵＩＤにより、一つのＣＯ２センサ１９０が特定できる。 The air conditioning controller 160 compares the acquired UUID with a database that associates pre-stored UUIDs with the CO2 sensor 190, and performs individual identification of the CO2 sensor 190 (step S202). One CO2 sensor 190 can be specified by the UUID.
　空調コントローラ１６０は、特定したＣＯ２センサ１９０単位で、得られたＣＯ２濃度情報を時刻とともに保存する（ステップＳ２０３）。 The air conditioning controller 160 stores the obtained CO2 concentration information along with the time for each identified CO2 sensor 190 (step S203).
　次に、空調コントローラ１６０は、ステップＳ２０１で受信したＣＯ２濃度情報が基準値を超えているかを判定する（ステップＳ２０４）。 Next, the air conditioning controller 160 determines whether the CO2 concentration information received in step S201 exceeds a reference value (step S204).
　ステップＳ２０４において、ＣＯ２濃度情報が基準値を超えていると判定された場合（ステップＳ２０４のＹＥＳ）、図８のＣＯ２濃度超過処理に移る（ステップＳ２０５）。 If it is determined in step S204 that the CO2 concentration information exceeds the reference value (YES in step S204), the process moves to the CO2 concentration excess process in FIG. 8 (step S205).
　一方、ステップＳ２０４において、受信したＣＯ２濃度情報が基準値を超えていないと判定された場合（ステップＳ２０４のＮＯ）、図７のＣＯ２濃度情報の蓄積処理を終了する。 On the other hand, if it is determined in step S204 that the received CO2 concentration information does not exceed the reference value (NO in step S204), the CO2 concentration information accumulation process of FIG. 7 is ended.
　図８は、実施形態１に係る空調コントローラ１６０のＣＯ２濃度超過処理のフローチャートである。 FIG. 8 is a flowchart of the CO2 concentration excess processing of the air conditioning controller 160 according to the first embodiment.
　図８に示すように、空調コントローラ１６０は、ＣＯ２濃度超過処理では、対象エリアの近くの換気装置１２０及び室内機１３０をＢＬＥ位置情報により特定する（ステップＳ３０１）。 As shown in FIG. 8, in the CO2 concentration excess process, the air conditioning controller 160 identifies the ventilation device 120 and indoor unit 130 near the target area using the BLE position information (step S301).
　次に、空調コントローラ１６０は、対象エリアの近くの換気装置１２０及び室内機１３０の稼働状態が、運転状態なのかを判断する（ステップＳ３０２）。 Next, the air conditioning controller 160 determines whether the ventilation device 120 and the indoor unit 130 near the target area are in the operating state (step S302).
　ステップＳ３０２において、換気装置１２０及び室内機１３０の稼働状態が、運転状態でないと判断された場合（ステップＳ３０２のＮＯ）、換気装置１２０及び室内機１３０の運転をＯＮにし、強制換気運転を行う（ステップＳ３０３）。 In step S302, if it is determined that the operating state of the ventilation device 120 and the indoor unit 130 is not in the operating state (NO in step S302), the operation of the ventilation device 120 and the indoor unit 130 is turned on and forced ventilation operation is performed ( Step S303).
　ステップＳ３０２において、換気装置１２０及び室内機１３０の稼働状態が、運転状態であると判断された場合（ステップＳ３０２のＹＥＳ）、換気装置１２０及び室内機１３０の対象エリア方向の風量を増加させる（ステップＳ３０４）。 In step S302, if it is determined that the operating state of the ventilation device 120 and the indoor unit 130 is in the operating state (YES in step S302), the air volume of the ventilation device 120 and the indoor unit 130 in the target area direction is increased (step S304).
　図９は、実施形態１に係る空調コントローラ１６０の推論処理のフローチャートである。 FIG. 9 is a flowchart of the inference process of the air conditioning controller 160 according to the first embodiment.
　空調コントローラ１６０は、推論開始が必要であるか否かを判断する（ステップＳ４０１）。具体的には、空調コントローラ１６０は、ＣＯ２濃度情報が、例えば、５００ｐｐｍの内部下限値を超えている場合に、推論開始が必要であると判断する。これは、低濃度状態では、換気の必要がないので、一定の下限値を設けて、ある程度ＣＯ２濃度情報が上昇したタイミングで、推論処理を開始する。内部下限値は、基準値及びＣＯ２濃度情報閾値よりも低い値である。入力パラメータとしては、ＣＯ２濃度情報、ＣＯ２濃度情報閾値及び在室人数が入力される。 The air conditioning controller 160 determines whether or not it is necessary to start inference (step S401). Specifically, the air conditioning controller 160 determines that it is necessary to start inference when the CO2 concentration information exceeds an internal lower limit of 500 ppm, for example. This is because in a low concentration state, there is no need for ventilation, so a certain lower limit value is set and the inference process is started at the timing when the CO2 concentration information has increased to a certain extent. The internal lower limit value is a value lower than the reference value and the CO2 concentration information threshold value. As input parameters, CO2 concentration information, CO2 concentration information threshold, and the number of people in the room are input.
　ステップＳ４０１において、ＣＯ２濃度情報の推論開始が必要でないと判断された場合（ステップＳ４０１のＮＯ）、図９の処理は終了する。 If it is determined in step S401 that it is not necessary to start inferring CO2 concentration information (NO in step S401), the process in FIG. 9 ends.
　ステップＳ４０１において、ＣＯ２濃度情報の推論開始が必要であると判断された場合（ステップＳ４０１のＹＥＳ）、ＣＯ２濃度情報の推論処理が行われる（ステップＳ４０２）。ステップＳ４０２では、図５に記載の入力層Ｉから出力層Ｏまでの推論処理が行われる。そして、ＣＯ２濃度情報がＣＯ２濃度情報閾値を超過するまでの時間、すなわち図４に示した到達時間が算出される。 If it is determined in step S401 that it is necessary to start inference of CO2 concentration information (YES in step S401), inference processing of CO2 concentration information is performed (step S402). In step S402, inference processing from the input layer I to the output layer O shown in FIG. 5 is performed. Then, the time until the CO2 concentration information exceeds the CO2 concentration information threshold, that is, the arrival time shown in FIG. 4 is calculated.
　次に、推論結果がＣＯ２濃度情報閾値を超過予定か否かが判断される（ステップＳ４０３）。ここでは、次にステップＳ４０１の処理が実施されるまでの時間よりも、ステップＳ４０２で得た到達時間の方が短い場合に、超過予定であると判断される。すなわち、周期的に実施される推論処理の次回のタイミングよりも早くに、ＣＯ２濃度がＣＯ２濃度情報閾値を超えるか否かを判断するのである。　 Next, it is determined whether the inference result is expected to exceed the CO2 concentration information threshold (step S403). Here, if the arrival time obtained in step S402 is shorter than the time until the next step S401 is executed, it is determined that the scheduled time has been exceeded. That is, it is determined whether the CO2 concentration exceeds the CO2 concentration information threshold earlier than the next timing of the periodically performed inference process.　
　ステップＳ４０３において、推論結果がＣＯ２濃度情報閾値を超過予定ではない場合（ステップＳ４０３のＮＯ）、図９の推論処理のフローチャートの処理は終了する。 In step S403, if the inference result is not expected to exceed the CO2 concentration information threshold (NO in step S403), the process of the inference process flowchart in FIG. 9 ends.
　ステップＳ４０３において、推論結果がＣＯ２濃度情報閾値を超過予定である場合（ステップＳ４０３のＹＥＳ）、図４で説明した強制動作時刻になると、対象エリアの換気装置１２０及び室内機１３０の強制稼働を行い（ステップＳ４０４）、処理を終了する。具体的には、ステップＳ４０４では、空調コントローラ１６０は、学習モデルｍｄにより推論された対象エリアのＣＯ２濃度情報がＣＯ２濃度情報閾値を超えている場合、ＢＬＥビーコンから発信されたＣＯ２センサ１９０の位置情報によって示される位置に対応する対象エリアの空調装置の換気を実施する。なお、換気装置１２０及び室内機１３０が既に稼働状態の場合は、換気装置１２０及び室内機１３０の風量を増加させる。 In step S403, if the inference result is scheduled to exceed the CO2 concentration information threshold (YES in step S403), at the forced operation time explained in FIG. (Step S404), the process ends. Specifically, in step S404, if the CO2 concentration information of the target area inferred by the learning model md exceeds the CO2 concentration information threshold, the air conditioning controller 160 uses the position information of the CO2 sensor 190 transmitted from the BLE beacon. Ventilate the air conditioner in the target area corresponding to the location indicated by. Note that when the ventilation device 120 and the indoor unit 130 are already in operation, the air volume of the ventilation device 120 and the indoor unit 130 is increased.
　図１０は、実施形態１に係る空調コントローラ１６０の学習処理のフローチャートである。 FIG. 10 is a flowchart of the learning process of the air conditioning controller 160 according to the first embodiment.
　図１０に示すように、空調コントローラ１６０は、学習開始時間であるか否かの判断を定期的に行う（ステップＳ５０１）。空調コントローラ１６０は、学習にはＣＰＵ負荷が増大するため、１日に一回、又は二回と予め決められた時刻に開始するようにする。 As shown in FIG. 10, the air conditioning controller 160 periodically determines whether it is learning start time (step S501). Since learning increases the CPU load, the air conditioning controller 160 starts learning at a predetermined time once or twice a day.
　次に、空調コントローラ１６０は、図５に記載の出力層Ｏから入力層Ｉへの学習処理をエリア単位で、実施する（ステップＳ５０２）。 Next, the air conditioning controller 160 performs the learning process from the output layer O to the input layer I described in FIG. 5 for each area (step S502).
　次に、空調コントローラ１６０は、学習結果に応じて、中間層Ｍ＿１、中間層Ｍ＿２及び中間層Ｍ＿３でのパラメータ、重み及びバイアス値を更新する（ステップＳ５０３）。空調コントローラ１６０は、記憶部である内部不揮発領域２４０に記憶された人情報及びＣＯ２濃度情報を基に、学習モデルｍｄの学習を行う。 Next, the air conditioning controller 160 updates the parameters, weights, and bias values in the intermediate layer M_1, intermediate layer M_2, and intermediate layer M_3 according to the learning results (step S503). The air conditioning controller 160 performs learning of the learning model md based on the person information and CO2 concentration information stored in the internal nonvolatile area 240, which is a storage unit.
　従って、実施形態１の空調システム１００によれば、学習モデルｍｄを使用して、対象エリアの推論されたＣＯ２濃度情報がＣＯ２濃度情報閾値を超えている場合、ＢＬＥビーコンから発信されたＣＯ２センサ１９０の位置情報によって示される位置に対応する対象エリアの空気調和を行うように空調装置を動作させる。これにより、対象エリアのＣＯ２濃度状態が基準値未満となるように、エリア毎に沿った適切な空気調和状態を維持することができる。 Therefore, according to the air conditioning system 100 of the first embodiment, when the CO2 concentration information inferred in the target area exceeds the CO2 concentration information threshold using the learning model md, the CO2 sensor 190 transmitted from the BLE beacon The air conditioner is operated to perform air conditioning in the target area corresponding to the position indicated by the position information. Thereby, it is possible to maintain an appropriate air conditioning state for each area so that the CO2 concentration state in the target area is less than the reference value.
　実施形態は、例として提示したものであり、請求の範囲を限定することは意図していない。実施形態は、その他の様々な形態で実施されることが可能であり、実施形態の要旨を逸脱しない範囲で、種々の省略、置き換え、変更を行うことができる。これら実施形態及びその変形は、実施形態の範囲及び要旨に含まれる。 The embodiments are presented as examples and are not intended to limit the scope of the claims. The embodiments can be implemented in various other forms, and various omissions, substitutions, and changes can be made without departing from the gist of the embodiments. These embodiments and variations thereof are included within the scope and spirit of the embodiments.
　１００　空調システム、１１０　室外機、１２０　換気装置、１３０　室内機、１４０　ＩｏＴゲートウェイ、１５０　人感センサ、１６０　空調コントローラ、１７０　外部メモリ、１８０　クラウド、１９０　ＣＯ２センサ、２００　ＢＬＥビーコン、２１０　ＣＯ２情報取得部、２２０　データ処理部、２３０　データ不揮発部、２４０　内部不揮発領域、２５０　外部不揮発領域、２６０　クラウド不揮発領域、３００　ＢＬＥセンサ情報取得部、３１０　ＣＯ２データ測定部、３２０　データ送受信部、Ｉ　入力相、Ｏ　出力層、Ｍ＿１、Ｍ＿２、Ｍ＿３　中間層、ｍｄ　学習モデル。 100 Air conditioning system, 110 Outdoor unit, 120 Ventilation system, 130 Indoor unit, 140 IoT gateway, 150 Human sensor, 160 Air conditioning controller, 170 External memory, 180 Cloud, 190 CO2 sensor, 200 BLE beacon, 210 CO2 information acquisition unit , 220 data processing unit, 230 data nonvolatile unit, 240 internal nonvolatile area, 250 external nonvolatile area, 260 cloud nonvolatile area, 300 BLE sensor information acquisition unit, 310 CO2 data measurement unit, 320 data transmission/reception unit, I input phase, O output layer , M_1, M_2, M_3 middle layer, md learning model.

Claims (5)

  1. 　屋内の空調設備である空調装置と、
    　前記空調装置を制御する空調コントローラと、
    　複数の対象エリアのそれぞれに設けられ、ＣＯ２濃度情報を取得するＣＯ２センサと、
    　前記ＣＯ２センサの位置を示す位置情報を発信するＢＬＥビーコンと
    を具備し、
    　前記空調コントローラは、
    　前記ＣＯ２センサにより取得された前記ＣＯ２濃度情報から前記対象エリア毎の将来の前記ＣＯ２濃度情報を推論する学習モデルを具備し、
    　前記学習モデルにより推論された前記対象エリアの前記将来のＣＯ２濃度情報が閾値を超える場合、前記ＢＬＥビーコンから発信された前記ＣＯ２センサの位置情報によって示される位置に対応する前記対象エリアの空気調和を行うように前記空調装置を動作させる
    空調システム。     Air conditioning equipment, which is indoor air conditioning equipment,
    an air conditioning controller that controls the air conditioning device;
    A CO2 sensor that is installed in each of the plurality of target areas and acquires CO2 concentration information;
    and a BLE beacon that transmits position information indicating the position of the CO2 sensor,
    The air conditioning controller includes:
    comprising a learning model that infers the future CO2 concentration information for each target area from the CO2 concentration information acquired by the CO2 sensor,
    If the future CO2 concentration information of the target area inferred by the learning model exceeds a threshold, air conditioning of the target area corresponding to the position indicated by the position information of the CO2 sensor transmitted from the BLE beacon is performed. An air conditioning system that operates said air conditioner to perform.
  2. 　前記対象エリア毎に前記ＣＯ２センサにより取得された現在及び過去の前記ＣＯ２濃度情報を蓄積する記憶部を具備する
    請求項１記載の空調システム。     The air conditioning system according to claim 1, further comprising a storage unit that stores the current and past CO2 concentration information acquired by the CO2 sensor for each target area.
  3. 　前記対象エリア毎に設けられ、前記対象エリアの人数と位置とを示す人情報を検出する人感センサを具備し、
    　前記記憶部は、前記対象エリア毎に、前記人感センサにより検出された人情報を記憶する
    請求項２記載の空調システム。     A human sensor is provided in each target area and detects human information indicating the number of people and positions in the target area,
    The air conditioning system according to claim 2, wherein the storage unit stores human information detected by the human sensor for each target area.
  4. 　前記空調コントローラは、前記記憶部に記憶された過去の前記人情報及び前記ＣＯ２濃度情報を基に、前記学習モデルの学習を行う
    請求項３記載の空調システム。     The air conditioning system according to claim 3, wherein the air conditioning controller performs learning of the learning model based on the past human information and the CO2 concentration information stored in the storage unit.
  5. 　前記空調装置は、換気を行う換気装置及び屋内に温度又は湿度が調節された調和空気の供給を行う室内機を含む
    請求項１～４のいずれか１項に記載の空調システム。     The air conditioning system according to any one of claims 1 to 4, wherein the air conditioning device includes a ventilation device that performs ventilation and an indoor unit that supplies conditioned air indoors with controlled temperature or humidity.
PCT/JP2022/011807 2022-03-16 2022-03-16 Air conditioning system WO2023175755A1 (en)

Priority Applications (2)

Application Number Priority Date Filing Date Title
PCT/JP2022/011807 WO2023175755A1 (en) 2022-03-16 2022-03-16 Air conditioning system
JP2024507281A JPWO2023175755A1 (en) 2022-03-16 2022-03-16

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/JP2022/011807 WO2023175755A1 (en) 2022-03-16 2022-03-16 Air conditioning system

Publications (1)

Publication Number Publication Date
WO2023175755A1 true WO2023175755A1 (en) 2023-09-21

Family

ID=88022532

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/JP2022/011807 WO2023175755A1 (en) 2022-03-16 2022-03-16 Air conditioning system

Country Status (2)

Country Link
JP (1) JPWO2023175755A1 (en)
WO (1) WO2023175755A1 (en)

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2009090906A (en) * 2007-10-11 2009-04-30 Denso Corp Co2 concentration detection system
JP2014134343A (en) * 2013-01-10 2014-07-24 Daikin Ind Ltd Air conditioning system
JP2018155461A (en) * 2017-03-21 2018-10-04 東芝キヤリア株式会社 Air conditioning management device
JP2020051723A (en) * 2018-09-28 2020-04-02 ダイキン工業株式会社 Abnormality determination device for transport refrigeration apparatus, transport refrigeration apparatus provided with abnormality determination device, and abnormality determination method for transportation refrigeration apparatus
JP2020094796A (en) * 2018-11-29 2020-06-18 ダイキン工業株式会社 Refrigerant leakage determination system and refrigeration cycle device

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2009090906A (en) * 2007-10-11 2009-04-30 Denso Corp Co2 concentration detection system
JP2014134343A (en) * 2013-01-10 2014-07-24 Daikin Ind Ltd Air conditioning system
JP2018155461A (en) * 2017-03-21 2018-10-04 東芝キヤリア株式会社 Air conditioning management device
JP2020051723A (en) * 2018-09-28 2020-04-02 ダイキン工業株式会社 Abnormality determination device for transport refrigeration apparatus, transport refrigeration apparatus provided with abnormality determination device, and abnormality determination method for transportation refrigeration apparatus
JP2020094796A (en) * 2018-11-29 2020-06-18 ダイキン工業株式会社 Refrigerant leakage determination system and refrigeration cycle device

Also Published As

Publication number Publication date
JPWO2023175755A1 (en) 2023-09-21

Similar Documents

Publication Publication Date Title
US10613555B2 (en) HVAC controller with wireless network based occupancy detection and control
US10837665B2 (en) Multi-function thermostat with intelligent ventilator control for frost/mold protection and air quality control
US10935270B2 (en) Electronic device and air-conditioning control method therefor
EP1698833B1 (en) Multi-air conditioner central control system
AU2017302766B2 (en) Multi air conditioner
CN101809514B (en) Application of microsystems for comfort control
JP4274163B2 (en) Air conditioning control system
EP1950505B1 (en) System for controlling demand of multi-air conditioner
JP2007107782A (en) Air conditioning control system and air conditioning control device
US20180320916A1 (en) Hvac management system and method
JP2007107871A (en) Facility control system and facility control device
WO2020075824A1 (en) Air conditioner, data transmission method, and air conditioning system
JP2006308280A (en) Air conditioning system and its control method
JP2002238081A (en) Control system and method for electrical appliance
US20200141597A1 (en) Determining window efficiency setting for use in humidity control of a building
US11333382B2 (en) System for heating, ventilation, air-conditioning
WO2023175755A1 (en) Air conditioning system
JP2020125868A (en) Air conditioning system
KR102085799B1 (en) Indoor air conditioning control system using environment measurement sensor connected to user terminal
JP7392394B2 (en) Air conditioning system and air conditioner
US10635067B2 (en) Intelligent building automation apparatus of autonomous control and control method thereof
KR102521766B1 (en) Air conditioning management system
US20230147059A1 (en) Air conditioning system, air-conditioning control program, and storage medium storing air-conditioning control program
US20230358424A1 (en) Operation of environmental control system during thermostat failure
JP2006329441A (en) Humidification air conditioning system and its program

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 22932035

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 2024507281

Country of ref document: JP

Kind code of ref document: A