air conditioning system

Abstract

To provide a technique capable of reducing discomfort that a user feels during reserved operation.SOLUTION: In an air conditioning system 1, a server device 5 generates a building load prediction model on the basis of operation information data acquired from an indoor unit 2, an adapter 3 predicts a building load by using the building load prediction model acquired from the server device 5 and determines advance operation start time on the basis of the predicted building load, and the indoor unit 2 acquires the advance operation start time determined by the adapter 3 from the adapter 3 and starts advance operation at the acquired advance operation start time.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present disclosure relates to air conditioning systems. [Background technology]

[0002] Air conditioning systems that utilize AI (Artificial Intelligence) are known. By utilizing AI in air conditioning systems, it is possible to realize a comfortable air-conditioned space that meets the user's preferences, behavioral patterns, etc. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Laid-Open No. 2015-117933 Summary of the Invention [Problem to be solved by the invention]

[0004] In air conditioning systems, in "reserved operation" where an air conditioner is started to operate later than the current time by a timer setting, the air conditioner is operated in advance (hereinafter sometimes referred to as "pre-operation") from a time before the reserved time until the reserved time so that the indoor temperature will be at the temperature set by the user (hereinafter sometimes referred to as the "user-set temperature") at the reserved time when the reserved operation starts. Below, the reservation to start operation of the air conditioner by setting a timer will be referred to as an "on timer reservation," and the start time of the reserved operation of the air conditioner will be referred to as the "on timer reservation time."

[0005] The "air conditioning load" (hereinafter sometimes referred to as "building load"), which is the energy required to maintain constant temperature and humidity in the room where the indoor unit is installed, differs from room to room. If the start time of pre-operation (hereinafter sometimes referred to as "pre-operation start time") is fixed to a predetermined time (for example, 40 minutes before the scheduled on-timer time), even if pre-operation starts before the scheduled on-timer time, if the building load is high, the indoor temperature may not reach the user-set temperature at the scheduled on-timer time, which may cause the user to feel uncomfortable.

[0006] The present disclosure provides a technology that can reduce the discomfort felt by users during scheduled driving. [Means for solving the problem]

[0007] In one embodiment of the disclosure, an air conditioning system includes an indoor unit, an adapter, and a server device. The indoor unit acquires from the adapter a first start time, which is the start time of a pre-operation that sets the indoor temperature to a user-set temperature at the scheduled on-timer time, and starts the pre-operation at the acquired first start time. The server device generates a learning model for predicting the building load of the room in which the indoor unit is installed based on operation information data acquired from the indoor unit. The adapter predicts the building load using the learning model acquired from the server device, and determines the first start time based on the predicted building load. [Effects of the Invention]

[0008] According to the disclosed aspects, it is possible to reduce the discomfort felt by the user when making a scheduled drive. [Brief explanation of the drawings]

[0009] [Figure 1] FIG. 1 is a diagram illustrating an example of the configuration of an air conditioning system according to a first embodiment. [Figure 2] FIG. 2 is a diagram illustrating an example of the configuration of the adapter according to the first embodiment. [Figure 3]FIG. 3 is a diagram illustrating an example of the configuration of the server device according to the first embodiment. [Figure 4] FIG. 4 is a diagram illustrating an example of driving information data according to the first embodiment. [Figure 5] FIG. 5 is a diagram illustrating an example of driving information data used to generate or update the sensible temperature setting prediction model according to the first embodiment. [Figure 6] FIG. 6 is a diagram illustrating an example of outdoor temperature forecast information according to the first embodiment. [Figure 7] FIG. 7 is a flowchart illustrating an example of processing performed by the air conditioning system of the first embodiment. [Figure 8] FIG. 8 is a diagram illustrating an example of the determination start time table according to the first embodiment. [Figure 9] FIG. 9 is a diagram illustrating an example of the preliminary operation start time determined by the AI ​​determination according to the first embodiment. [Figure 10] FIG. 10 is a diagram illustrating an example of the normal determination table according to the first embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0010] The technology of the present disclosure will be described below with reference to the drawings. In the following, the same components are denoted by the same reference numerals.

[0011] [Example 1] <Air conditioning system configuration> Fig. 1 is a diagram showing an example of the configuration of an air conditioning system of Example 1. In Fig. 1, the air conditioning system 1 has an indoor unit 2, an adapter 3, routers 4A and 4B, a server device 5, a relay device 6, a communication terminal 7, and a communication network 8. The adapter 3 and the relay device 6 can communicate with each other via the router 4A and the communication network 8. Furthermore, the communication terminal 7 and the relay device 6 can communicate with each other via the router 4B and the communication network 8.

[0012] In the following, the judgment made by AI using a learning model may be referred to as "AI judgment," and controlling the air conditioner using the results of AI judgment may be referred to as "AI control."

[0013] The indoor unit 2 is a part of an air conditioner that is placed indoors and heats or cools the air inside the room. The air conditioner mainly has the indoor unit 2 and an outdoor unit (not shown) placed outdoors. A user of the air conditioner can remotely control the indoor unit 2 by operating a remote control 9. Examples of the remote control 9 include an infrared remote control or a radio remote control. The indoor unit 2 has a main body 2A and a control unit 2B that controls the main body 2A. The control unit 2B is realized by a processor such as a CPU (Central Processing Unit), an MPU (Micro Processing Unit), or an MCU (Micro Controller Unit). The main body 2A is equipped with an indoor fan and an indoor heat exchanger, etc., and indoor air that has exchanged heat with a refrigerant in the indoor heat exchanger is blown out from the main body 2A to heat, cool, or dehumidify the room. The main body 2A is also provided with operation buttons that can directly operate the indoor unit 2. The outdoor unit, which is another part of the air conditioner, is equipped with an outdoor fan, an outdoor heat exchanger, a compressor, an expansion valve, etc.

[0014] The adapter 3 has a communication function for connecting the indoor unit 2 and the router 4A via wireless communication, and a control function for controlling the indoor unit 2 using AI (Artificial Intelligence). An adapter 3 is installed for each indoor unit 2. The router 4A connects to the adapter 3 via wireless communication, for example, using a WLAN (Wireless Local Area Network), and connects the adapter 3 to a communication network 8. An example of the communication network 8 is the Internet.

[0015] Examples of the communication terminal 7 include a smartphone, a tablet terminal, etc. used by a user. The router 4B connects to the communication terminal 7 via wireless communication, for example, using WLAN, etc., and connects the communication terminal 7 to the communication network 8. The user can operate the indoor unit 2 by operating the communication terminal 7.

[0016] The server device 5 has a function to generate a learning model for AI that controls the indoor units 2, and a database that stores operation information data of the air conditioner. The server device 5 is installed in, for example, a data center. The relay device 6 is connected to a communication network 8 and has a function to communicate with the server device 5. The relay device 6 receives operation information data used to generate or update the learning model applied to the adapter 3 from the adapter 3, and transmits the received operation information data to the server device 5. The relay device 6 also receives from the server device 5 a learning model generated or updated by the server device 5, and transmits the received learning model to the adapter 3.

[0017] The relay device 6 has a first relay unit 6A, a second relay unit 6B, and a third relay unit 6C. The first relay unit 6A transmits various data related to AI control between the adapter 3 and the server device 5. For example, the first relay unit 6A transmits operation information data received from the adapter 3 to the server device 5, and transmits a learning model generated or updated by the server device 5 using the operation information data to the adapter 3. The second relay unit 6B acquires the operation conditions (such as the operation mode, e.g., cooling / heating, and the set temperature) of the indoor unit 2 set by the user using the communication terminal 7, and transmits the acquired operation conditions to the indoor unit 2 via the adapter 3. The third relay unit 6C acquires external data, such as a weather forecast, via the communication network 8 and transmits the acquired external data to the server device 5 and the adapter 3.

[0018] <Adapter configuration> 2 is a diagram showing an example of the configuration of the adapter of Example 1. In FIG. 2, the adapter 3 has a first communication unit 11, a second communication unit 12, a storage unit 13, and a processor .

[0019] The first communication unit 11 communicates with the control unit 2B of the indoor unit 2 and is realized by a communication IF (Interface) such as a UART (Universal Asynchronous Receiver Transmitter). The second communication unit 12 communicates with the router 4A and is realized by a communication IF for WLAN, for example. The storage unit 13 is realized by a HDD (Hard Disk Drive), ROM (Read Only Memory), RAM (Random Access Memory), etc., and stores various information such as data and programs. The processor 14 is realized by a CPU, MPU, MCU, etc.

[0020] The storage unit 13 has an operation information memory 13A, a model memory 13B, and an external memory 13C. The operation information memory 13A temporarily stores operation information data acquired from the indoor units 2. The model memory 13B stores learning models acquired from the server device 5. The external memory 13C stores external data.

[0021] The processor 14 has, as functional blocks thereof, an acquisition unit 14A, a transmission unit 14B, a reception unit 14C, a setting unit 14D, and a prediction unit 14E.

[0022] The acquisition unit 14A acquires operation information data such as the set temperature and room temperature from the indoor unit 2 at a predetermined cycle (for example, every 5 minutes). The acquisition unit 14A stores the acquired operation information data in the operation information memory 13A. As will be described later with reference to FIG. 4, the operation information data includes a timestamp that indicates the date on which the operation information data was acquired.

[0023] The transmission unit 14B acquires the driving information data stored in the driving information memory 13A from the driving information memory 13A, and transmits the acquired driving information data to the server device 5.

[0024] The receiving unit 14C receives the learning model from the server device 5, and stores the received learning model in the model memory 13B.

[0025] The setting unit 14D acquires the learning model stored in the model memory 13B from the model memory 13B, and sets the acquired learning model in the prediction unit 14E.

[0026] The prediction unit 14E uses the learning model set by the setting unit 14D to control the control unit 2B of the indoor unit 2. The prediction unit 14E may also use the learning model to directly control the main body 2A of the indoor unit 2. The prediction unit 14E may also indirectly control the main body 2A via the control unit 2B by transmitting a control mode based on the learning model to the control unit 2B.

[0027] <Server device configuration> Fig. 3 is a diagram illustrating an example of the configuration of a server device according to the first embodiment. In Fig. 3, the server device 5 includes a communication unit 31, a storage unit 32, and a processor 33. The communication unit 31 communicates with the relay device 6 and is realized, for example, by a communication IF. The storage unit 32 is realized, for example, by an HDD, a ROM, a RAM, etc., and stores various information such as data and programs. The processor 33 is realized, for example, by a CPU, an MPU, an MCU, etc.

[0028] The storage unit 32 has a data memory 32A and a model memory 32B. The data memory 32A stores the driving information data received from the adapter 3. The model memory 32B stores the learning model generated or updated by the server device 5.

[0029] The processor 33 has, as functional blocks thereof, a receiving unit 33A, a learning unit 33B, and a transmitting unit 33C.

[0030] The receiving unit 33A receives operation information data from each adapter 3 connected to each of the indoor units 2, and stores the received operation information data in the data memory 32A.

[0031] The learning unit 33B performs machine learning using the operation information data stored in the data memory 32A, and generates or updates a learning model based on the learning results. The learning unit 33B stores the generated or updated learning model in the model memory 32B. One example of a learning model is a "sensible temperature setting prediction model" that predicts the user's sensible temperature based on the operating status of each home's air conditioner and controls the air conditioner according to the predicted sensible temperature.

[0032] The transmission unit 33C acquires the learning model stored in the model memory 32B from the model memory 32B, and transmits the acquired learning model to the adapter 3 via the relay device 6.

[0033] <Example of driving information data> 4 is a diagram showing an example of operation information data according to Example 1. The operation information data includes, for example, an operation state, an operation mode, a set temperature, an indoor temperature (room temperature), an indoor humidity, an air volume, a wind direction, a human presence sensor, a radiation sensor, an indoor heat exchanger temperature, an outdoor temperature (outdoor air temperature), a compressor rotation speed, an outdoor air volume, an operating current, an outdoor heat exchanger temperature, a discharge temperature, a compressor temperature, an expansion valve opening, a radiator temperature, a start-up failure history, an abnormal stop history, an emergency operation history, a time stamp, an air conditioner ID, an installation location, a facility type, and the like.

[0034] The operating state indicates the ON / OFF state of the indoor unit 2. The operating mode indicates the operating mode of the indoor unit 2, such as cooling or heating. The set temperature is a temperature set by the user and indicates the target temperature of the room where the indoor unit 2 is used. The indoor temperature indicates the actual temperature of the room where the indoor unit 2 is used. The indoor humidity indicates the actual humidity of the room where the indoor unit 2 is used. The air volume indicates the air volume of the indoor air blown out from the indoor unit 2. The air direction indicates the air direction of the indoor air blown out from the indoor unit 2. The human presence sensor indicates the detection results of the sensor on the presence or absence of people in the room and the amount of activity. The radiation sensor indicates the detection results of the temperature of the indoor floor and walls. The indoor heat exchanger temperature indicates the temperature of the indoor heat exchanger that is part of the main body 2A of the indoor unit 2. The outdoor temperature indicates the actual outdoor temperature. The compressor rotation speed indicates the operating rotation speed of the compressor provided in the outdoor unit connected to the indoor unit 2 by refrigerant piping. The outdoor air volume indicates the air volume generated by the outdoor fan provided in the outdoor unit. The operating current indicates the operating current of the entire air conditioner, for example, the indoor unit 2 and outdoor unit. The outdoor heat exchanger temperature indicates the temperature of the outdoor heat exchanger provided in the outdoor unit. The discharge temperature indicates the temperature of the refrigerant discharged from the compressor. The compressor temperature indicates the temperature at the bottom of the compressor. The expansion valve opening indicates the opening of the electronic expansion valve provided in the outdoor unit. The radiator temperature indicates the temperature of the power semiconductor that drives and controls the compressor. The start-up failure history indicates the history of compressor start-up failures. The abnormal stop history indicates the history of abnormal stoppages of the air conditioner. The emergency operation history indicates the history of emergency operation. The timestamp indicates the acquisition date and time of each operating information data in year, month, day, hour, minute, and second. The air conditioner ID indicates an ID assigned to the indoor unit 2 to identify the air conditioner. The installation location indicates the address of the location where the air conditioner is installed. The facility type indicates the type of facility (store, restaurant, factory, etc.) in which the air conditioner is installed.

[0035] The various pieces of operational information data shown in Figure 4 are used depending on the purpose of the air conditioner, such as for home or commercial use. Examples of operational information data used for home air conditioners include the operating status, operating mode, set temperature, indoor temperature, indoor humidity, air volume, wind direction, human presence sensor, radiation sensor, timestamp, air conditioner ID, and installation location. Home air conditioners use AI to perform operations and make suggestions in pursuit of comfort and energy efficiency, so data necessary for home use includes, for example, the set temperature, operating mode, and indoor and surrounding environment.

[0036] On the other hand, operational information data used for commercial air conditioners includes, for example, operational status, operational mode, set temperature, indoor temperature, indoor humidity, air volume, wind direction, motion sensor, radiation sensor, indoor heat exchanger temperature, outdoor temperature, compressor rotation speed, outdoor air volume, operating current, outdoor heat exchanger temperature, discharge temperature, compressor temperature, expansion valve opening, radiator temperature, start-up failure history, abnormal shutdown history, emergency operation history, timestamp, air conditioner ID, installation location, and facility type. In commercial air conditioners, AI predicts the failure and maintenance needs of each device. For example, in commercial air conditioners, the operating status and history of each component within the air conditioner are accumulated, and the AI ​​predicts the timing of component failure based on the accumulated operating status and history. Note that operational information data is not generated for the compressor and fan motor installed in the air conditioner when the air conditioner is stopped. Therefore, data such as compressor rotation speed, outdoor air volume, operating current, and outdoor heat exchanger temperature do not need to be acquired when the air conditioner is stopped.

[0037] When the learning model is, for example, a sensible temperature setting prediction model, time-series driving information data such as the set temperature, indoor temperature, indoor humidity, and outdoor temperature are used to generate or update the sensible temperature setting prediction model. FIG. 5 is a diagram showing an example of driving information data used to generate or update the sensible temperature setting prediction model of the first embodiment. As shown in FIG. 5, the driving information data used in the sensible temperature setting prediction model differs depending on the season. For example, a winter sensible temperature setting prediction model uses the set temperature, indoor temperature, indoor humidity, and outdoor temperature. On the other hand, a summer sensible temperature setting prediction model uses, in addition to the driving information data used in winter, for example, air volume and detection data from a human presence sensor (presence or absence of people and activity level).

[0038] <Air conditioning system processing and operation> When an air conditioner is operating in cooling mode, if the outdoor temperature is the same, the smaller the building load and the higher the air conditioning capacity currently being exerted by the air conditioner (hereinafter sometimes referred to as "current capacity"), the faster the indoor temperature will drop. Also, when an air conditioner is operating in cooling mode, if the outdoor temperature is the same, the smaller the absolute value of the difference between the indoor temperature and a user-set temperature that is lower than the indoor temperature, the faster the indoor temperature will reach the user-set temperature. Also, when an air conditioner is stopped after operating in cooling mode, if the outdoor temperature is the same, the higher the building load, the faster the indoor temperature will rise. Also, when an air conditioner is stopped after operating in cooling mode, if the building load is the same, the higher the outdoor temperature, the faster the indoor temperature will rise.

[0039] On the other hand, when an air conditioner is operating in heating mode, if the outdoor temperature is the same, the smaller the building load and the higher the current capacity, the faster the indoor temperature will rise. Also, when an air conditioner is operating in heating mode, if the outdoor temperature is the same, the smaller the absolute value of the difference between the indoor temperature and a user-set temperature that is higher than the indoor temperature, the faster the indoor temperature will reach the user-set temperature. Also, when an air conditioner is stopped after operating in heating mode, if the outdoor temperature is the same, the higher the building load, the faster the indoor temperature will drop. Also, when an air conditioner is stopped after operating in heating mode, if the building load is the same, the lower the outdoor temperature, the faster the indoor temperature will drop.

[0040] Therefore, when the learning unit 33B of the server device 5 generates or updates a learning model for predicting the building load of a room in which the indoor unit 2 is installed (hereinafter, sometimes referred to as the "building load prediction model"), the adapter 3 acquires the current performance capacity from the indoor unit 2 at a predetermined interval (for example, every five minutes) as operation information data in addition to the various pieces of information shown in FIG. 4, and transmits the acquired current performance capacity to the server device 5. The learning unit 33B performs machine learning using the current performance capacity and the outdoor temperature and indoor temperature in FIG. 4 as operation information data, and generates or updates the building load prediction model based on the learning results. The generated or updated building load prediction model is transmitted from the server device 5 to the adapter 3 by the transmission unit 33C and stored in the model memory 13B. The building load prediction model stored in the model memory 13B is set in the prediction unit 14E by the setting unit 14D.

[0041] 7 is a flowchart illustrating an example of processing performed by the air conditioning system of Example 1. In the following, time is expressed in 24-hour format.

[0042] The flowchart shown in Figure 7 is started at a predetermined interval (for example, every 5 minutes) only when the AI ​​control is set to on by the user. In other words, pre-operation is performed only when the AI ​​control is set to on by the user, and when the AI ​​control is set to off by the user, scheduled operation is possible but pre-operation is not performed.

[0043] 7, in step S101, the control unit 2B of the indoor unit 2 determines whether or not an on-timer reservation has been set. If an on-timer reservation has been set (step S101: Yes), the control unit 2B sends the on-timer reservation time to the adapter 3, and the process proceeds to step S103. The on-timer reservation time sent from the control unit 2B is input to the prediction unit 14E via the first communication unit 11 of the adapter 3. On the other hand, if an on-timer reservation has not been set (step S101: No), the processes of steps S103 to S123 are not performed, and the process ends.

[0044] In step S103, control unit 2B waits until the current time is six minutes before the start time of AI determination for predicting the building load (hereinafter sometimes referred to as the "AI determination start time") (step S103: No). As described above, because adapter 3 and indoor unit 2 communicate every five minutes, in the processes of steps S105 and S107 below, in order to reliably obtain the indoor temperature detected after driving the indoor fan for one minute by the AI ​​determination start time, the process of step S105 is performed six minutes before the AI ​​determination start time. If the current time is six minutes before the AI ​​determination start time (step S103: Yes), the process proceeds to step S105.

[0045] Here, the control unit 2B makes the determination in step S103 in accordance with the determination start time table TA1 shown in FIG. 8. FIG. 8 is a diagram showing an example of the determination start time table of the first embodiment. The determination start time table TA1 is pre-stored in a memory (not shown) of the indoor unit 2. As shown in FIG. 8, the determination start time table TA1 pre-stores a correspondence between the operation mode immediately before the operation of the air conditioner is stopped (hereinafter, may be referred to as the "operation mode before operation stop") and the AI ​​determination start time. In step S103, if the operation mode before operation stop is "cooling," the control unit 2B determines from the determination start time table TA1 that the AI ​​determination start time is 40 minutes before the scheduled on-timer time. Therefore, for example, if the operation mode before operation stop is "cooling" and the scheduled on-timer time is 6:00, the control unit 2B determines the AI ​​determination start time to be 5:20. Furthermore, when the operation mode before operation shutdown is "heating," control unit 2B determines from judgment start time table TA1 that the AI ​​judgment start time is 60 minutes before the scheduled on-timer time. Therefore, for example, when the operation mode before operation shutdown is "heating" and the scheduled on-timer time is 6:00, control unit 2B determines the AI ​​judgment start time to be 5:00. The operation mode before operation shutdown is stored in advance in the memory of indoor unit 2 when operation of indoor unit 2 ends during the setting of the scheduled on-timer.

[0046] Returning to FIG. 7, in step S105, the control unit 2B drives the indoor fan of the indoor unit 2 for one minute in advance to detect the indoor temperature.

[0047] After driving the indoor fan for one minute, the control unit 2B detects the current indoor temperature (hereinafter may be referred to as the "current indoor temperature") in step S107. The control unit 2B also calculates the air conditioning capacity (hereinafter may be referred to as the "start-up capacity") to be exerted by the air conditioner when the air conditioner starts operating, based on the difference between the current indoor temperature and the user-set temperature. Typically, the greater the difference between the current indoor temperature and the user-set temperature, the higher the start-up capacity required. The control unit 2B transmits the detected current indoor temperature and the calculated start-up capacity to the adapter 3. The current indoor temperature and the start-up capacity are input to the prediction unit 14E via the first communication unit 11 of the adapter 3.

[0048] Next, in step S109, control unit 2B waits until five minutes have passed since performing the process of step S107 (step S109: No), and after five minutes have passed since performing the process of step S107 (step S109: Yes), the process proceeds to step S111. As will be described later, prediction unit 14E of adapter 3 obtains outdoor temperature forecast information (hereinafter sometimes referred to as "outdoor temperature forecast information") every five minutes, so control unit 2B waits until five minutes have passed since performing the process of step S107.

[0049] In step S111, the prediction unit 14E determines the preliminary operation start time by AI determination.

[0050] That is, first, the prediction unit 14E acquires outdoor temperature forecast information from outside the air conditioning system 1 via the relay device 6 and the communication network 8. Outdoor temperature is also called "outdoor air temperature." The outdoor temperature forecast information is included, for example, in a weather forecast, which is one of the external data acquired by the third relay unit 6C through the communication network 8. The prediction unit 14E acquires outdoor temperature forecast information, for example, every five minutes. For example, when the current date and time is between 18:00 and 18:59 on December 24, 2018, the prediction unit 14E acquires outdoor temperature forecast information FI shown in FIG. 6. FIG. 6 is a diagram illustrating an example of outdoor temperature forecast information according to the first embodiment. The outdoor temperature forecast information FI (FIG. 6) includes outdoor temperature forecast values ​​for 24 hours, measured every hour, starting from 18:00 on December 24, 2018. In FIG. 6, for example, the outdoor temperature forecast value of 6°C at 6:00 PM on December 24, 2018 indicates the outdoor temperature forecast value from 6:00 PM to 6:59 PM on December 24, 2018. Furthermore, prediction unit 14E extracts the forecast value of the outdoor temperature at the on-timer reservation time (hereinafter sometimes referred to as the "outdoor temperature forecast value") from outdoor temperature forecast information FI. For example, if the on-timer reservation time is 6:00 AM, prediction unit 14E extracts the outdoor temperature forecast value FV (0°C) at the on-timer reservation time of 6:00 AM from outdoor temperature forecast information FI (FIG. 6). In this way, prediction unit 14E extracts the latest outdoor temperature forecast value at the on-timer reservation time from the outdoor temperature forecast information.

[0051] Next, the prediction unit 14E inputs the current indoor temperature and start-up capacity obtained from the control unit 2B, and the extracted outdoor temperature forecast value (i.e., the outdoor temperature forecast value at the scheduled on-timer time), into a building load prediction model, and uses the building load prediction model to predict the building load of the room in which the indoor unit 2 is installed.

[0052] Then, the prediction unit 14E determines the preliminary operation start time based on the predicted building load.

[0053] FIG. 9 is a diagram illustrating an example of a pre-operation start time determined by AI determination in the first embodiment. As shown in FIG. 9, for example, when the operation mode before operation shutdown is "cooling," the prediction unit 14E determines the pre-operation start time in 5-minute increments within a range of 0 to 40 minutes before the scheduled on-timer time based on the predicted building load. For example, when the operation mode before operation shutdown is "heating," the prediction unit 14E determines the pre-operation start time in 5-minute increments within a range of 0 to 60 minutes before the scheduled on-timer time based on the predicted building load. For example, the pre-operation start time 30 minutes before the scheduled on-timer time of 6:00 is 5:30. The pre-operation start time determined by the prediction unit 14E is usually earlier than the scheduled on-timer time as the predicted building load increases. The prediction unit 14E transmits the determined pre-operation start time to the control unit 2B of the indoor unit 2 using the first communication unit 11. Communication between the adapter 3 and the control unit 2B occurs every 5 minutes.

[0054] For example, in Figure 9, the AI-determination start time "40 minutes before" when the operation mode before shutdown is "cooling" corresponds to the maximum expected building load during a specified summer period (hereinafter referred to as the "specified summer period") when the air conditioner is operating in cooling mode, and the AI-determination start time "0 minutes before" when the operation mode before shutdown is "cooling" corresponds to the minimum expected building load during the specified summer period. Furthermore, for example, the AI-determination start time "60 minutes before" when the operation mode before shutdown is "heating" corresponds to the maximum expected building load during a specified winter period (hereinafter referred to as the "specified winter period") when the air conditioner is operating in cooling mode, and the AI-determination start time "0 minutes before" when the operation mode before shutdown is "heating" corresponds to the minimum expected building load during the specified winter period.

[0055] Returning to FIG. 7 , next, in step S113, the control unit 2B determines whether or not the acquisition of the preliminary operation start time from the adapter 3 was successful. For example, if the prediction unit 14E fails to communicate with the outside of the air conditioning system 1 and is unable to acquire outdoor temperature forecast information in step S111, the prediction unit 14E is unable to determine the preliminary operation start time through AI determination, and the control unit 2B is therefore unable to acquire the preliminary operation start time from the prediction unit 14E. Therefore, for example, if the control unit 2B is able to acquire the preliminary operation start time from the prediction unit 14E, the control unit 2B determines that the acquisition of the preliminary operation start time was successful, and if the control unit 2B is unable to acquire the preliminary operation start time from the prediction unit 14E, the control unit 2B determines that the acquisition of the preliminary operation start time was unsuccessful. In addition to when the prediction unit 14E is unable to acquire outdoor temperature forecast information from the outside of the air conditioning system 1, the prediction unit 14E is unable to determine the preliminary operation start time through AI determination, for example, if the predicted building load is a value outside a predetermined range. If the control unit 2B succeeds in acquiring the pre-operation start time from the adapter 3 (step S113: Yes), the process proceeds to step S115; if the control unit 2B fails to acquire the pre-operation start time from the adapter 3 (step S113: No), the process proceeds to step S117.

[0056] In step S117, control unit 2B determines whether the current time has reached the start time (hereinafter sometimes referred to as the "normal determination start time") of determination by control unit 2B (hereinafter sometimes referred to as the "normal determination"). The normal determination is a determination made when the preliminary operation start time cannot be acquired in step S113, and in the normal determination, the preliminary operation start time is determined according to FIG. 10 when a predetermined normal determination start time (FIG. 8) has arrived. If the current time has not reached the normal determination start time (step S117: No), the process returns to step S113. On the other hand, if the current time has reached the normal determination start time (step S117: Yes), the process proceeds to step S119.

[0057] Here, control unit 2B makes the determination in step S117 in accordance with determination start time table TA1 shown in FIG. 8. As shown in FIG. 8, determination start time table TA1 has preset associations between operation modes before operation shutdown and normal determination start times. In step S117, if the operation mode before operation shutdown is "cooling," control unit 2B determines from determination start time table TA1 that the normal determination start time is 20 minutes before the scheduled on-timer time. Therefore, for example, if the operation mode before operation shutdown is "cooling" and the scheduled on-timer time is 6:00, control unit 2B determines that the normal determination start time is 5:40. Furthermore, if the operation mode before operation shutdown is "heating," control unit 2B determines from determination start time table TA1 that the normal determination start time is 45 minutes before the scheduled on-timer time. Therefore, for example, if the operation mode before operation shutdown is "heating" and the reserved time for the on timer is 6:00, control unit 2B determines the normal determination start time to be 5:15.

[0058] Returning to FIG. 7, in step S119, the control unit 2B determines the preliminary operation start time through normal determination. That is, in step S119, the control unit 2B determines the preliminary operation start time in accordance with normal determination table TA2 shown in FIG. 10, based on the operation mode before operation shutdown, the temperature set by the user, and the current indoor temperature detected in step S107. FIG. 10 is a diagram showing an example of the normal determination table of the first embodiment. The normal determination table TA2 is pre-stored in memory provided in the indoor unit 2. As shown in FIG. 10, the normal determination table TA2 pre-sets associations between the absolute value of the difference between the temperature set by the user and the current indoor temperature (hereinafter sometimes referred to as the "absolute temperature difference"), the operation mode before operation shutdown, and the preliminary operation start time.

[0059] For example, if the operation mode before operation shutdown is "cooling" and the absolute value of the temperature difference is equal to or greater than 5°C and less than 10°C, control unit 2B determines the pre-operation start time to be 15 minutes before the scheduled on-timer time in accordance with normal determination table TA2. Therefore, for example, if the scheduled on-timer time is 6:00, control unit 2B determines the pre-operation start time to be 5:45, 15 minutes before 6:00.

[0060] Furthermore, for example, if the operation mode before operation shutdown is "heating" and the absolute value of the temperature difference is 15°C or more and less than 20°C, control unit 2B determines the pre-operation start time to be 30 minutes before the scheduled on-timer time in accordance with normal determination table TA2. Therefore, for example, if the scheduled on-timer time is 6:00, control unit 2B determines the pre-operation start time to be 5:30, which is 30 minutes before 6:00.

[0061] That is, if the controller 2B cannot obtain the preliminary operation start time from the adapter 3 (step S113: No), the controller 2B uses the preliminary operation start time that is set and stored in advance (step S119).

[0062] If the process proceeds from step S113 to step S115, in step S115, the control unit 2B starts the preliminary operation at the preliminary operation start time determined in step S111 by the AI ​​determination of the prediction unit 14E. On the other hand, if the process proceeds from step S119 to step S115, in step S115, the control unit 2B starts the preliminary operation at the preliminary operation start time determined by the control unit 2B in step S119.

[0063] Next, in step S121, control unit 2B continues the preliminary operation until the current time reaches the reserved on-timer time (step S121: No). If the current time reaches the reserved on-timer time (step S121: Yes), the process proceeds to step S123.

[0064] In step S123, the control unit 2B ends the preliminary operation and starts reserved operation of the air conditioner in the operation before shutdown mode. After the processing of step S123, the processing ends.

[0065] As described above, in Example 1, the air conditioning system 1 has the indoor units 2, the adapter 3 that controls the indoor units 2 using AI, and the server device 5. The server device 5 generates a building load prediction model based on operation information data acquired from the indoor units 2. The adapter 3 predicts the building load using the building load prediction model acquired from the server device 5, and determines a pre-operation start time based on the predicted building load. The indoor unit 2 acquires the pre-operation start time determined by the adapter 3 from the adapter 3, and starts pre-operation at the acquired pre-operation start time. The pre-operation start time determined by the adapter 3 corresponds to the "first start time."

[0066] In this way, pre-operation is started at the pre-operation start time adjusted according to the building load, so even if the building load is large, the room temperature will surely reach the temperature set by the user at the on-timer reserved time, thereby reducing the discomfort felt by the user during reserved operation.

[0067] Furthermore, in the first embodiment, if the indoor unit 2 cannot obtain the preliminary operation start time from the adapter 3, it adopts a preliminary operation start time that has been set and stored in advance, and starts preliminary operation at the adopted preliminary operation start time. The preliminary operation start time that has been set and stored in advance corresponds to the "second start time."

[0068] By doing so, even if it is difficult to determine the preliminary operation start time by AI determination, the indoor unit 2 can acquire the preliminary operation start time.

[0069] In the first embodiment, the indoor unit 2 has an indoor fan. The indoor temperature is included in the operation information data. The indoor unit 2 drives the indoor fan before detecting the indoor temperature used to predict the building load in the adapter 3.

[0070] In this way, the indoor temperature is detected while the indoor air is stirred, allowing for accurate detection of the indoor temperature, thereby improving the accuracy of building load predictions.

[0071] The first embodiment has been described above.

[0072] [Example 2] In Example 1, the adapter 3 transmits the operation information data of the indoor unit 2 to the server device 5 via the relay device 6 as an example, but the adapter 3 may also transmit the operation information data directly to the server device 5 without going through the relay device 6.

[0073] Furthermore, the components shown in the figures do not necessarily have to be physically configured as shown in the figures. The specific form of distribution and integration of the components is not limited to that shown in the figures, and all or part of the components can be functionally or physically distributed and integrated in any unit depending on, for example, various loads and usage conditions.

[0074] Furthermore, all or part of the processes described above in the control unit 2B or the adapter 3 may be realized by having a processor in the control unit 2B or the adapter 3 execute a program corresponding to each process. For example, the programs corresponding to each process described above may be stored in memory, and the programs may be read from the memory and executed by the processor. Alternatively, the programs may be stored in a program server connected to the control unit 2B or the adapter 3 via any network, and downloaded from the program server to the control unit 2B or the adapter 3 for execution.

[0075] The second embodiment has been described above. [Explanation of symbols]

[0076] 1. Air conditioning system 2 Indoor unit 3 Adapters 5. Server equipment

Claims

1. An air conditioning system having an air conditioner with an indoor unit, an adapter, and a server device, The indoor unit detects the indoor temperature before the scheduled on-timer time, calculates an operation start capacity which is the air conditioning capacity to be exerted by the air conditioner when the operation of the air conditioner starts using the temperature difference between the detected indoor temperature and the set temperature, obtains from the adapter a first start time which is the start time of a pre-operation to bring the indoor temperature to the set temperature at the scheduled on-timer time, and starts the pre-operation at the obtained first start time, the server device generates a learning model for predicting the building load of the room in which the indoor unit is installed, based on the operation information data acquired from the indoor unit; the adapter acquires the operation start capacity detected and calculated by the indoor unit, uses the operation start capacity as an input value, predicts the building load using the learning model acquired from the server device, and determines the first start time based on the predicted building load. Air conditioning system.

2. the adapter acquires the indoor temperature detected by the indoor unit, and predicts the building load using the learning model with the indoor temperature as an input value in addition to the operation start capacity. The air conditioning system of claim 1 .

3. the adapter uses the indoor temperature, the operation start capacity, and an outdoor temperature acquired from outside as input values, and predicts the building load using the learning model. The air conditioning system according to claim 2 .

4. The outdoor temperature is the temperature predicted at the scheduled time of the on timer. The air conditioning system according to claim 3 .

5. When the first start time cannot be obtained from the adapter, the indoor unit adopts a second start time that is set and stored in advance, and starts the preliminary operation at the adopted second start time. The air conditioning system of claim 1 .

6. The indoor unit has an indoor fan, the indoor unit drives the indoor fan before detecting the indoor temperature used for predicting the building load; The air conditioning system according to claim 2 .

7. The indoor unit changes the determination start time depending on the operation mode immediately before the operation of the air conditioner is stopped. The air conditioning system according to claim 6.

8. The indoor unit further changes the time at which the indoor temperature is detected depending on the operation mode. The air conditioning system according to claim 7.

Images