Air conditioning system and learning device

Abstract

An air conditioning system (1) is provided with: an air conditioning apparatus (10) that has an air conditioning unit (13) for performing air conditioning operation and a control unit (11) for controlling the air conditioning unit (13); an inference data acquisition unit that acquires inference data (38) including an air conditioning environment (31), which is information about the environment in which the air conditioning apparatus (10) is installed; an inference device (35) that infers, from the air conditioning environment (31), proposal settings (39) that include at least two settings that match the manipulation preference of a user on the air conditioning apparatus (10); and a communication terminal (20) that remotely manipulates the air conditioning apparatus (10). The communication terminal (20) is provided with a display unit that displays a plurality of icons representing functions of the air conditioning apparatus (10). The control unit (11) causes the display unit to display proposal setting icons for performing the proposal settings inferred by the inference device (35), and when a manipulation of pressing a proposal setting icon is performed, the control unit (11) controls the air conditioning apparatus (10) in accordance with a proposal setting corresponding to the proposal setting icon.

Description

Air conditioning system and learning device

[0001] The present disclosure relates to an air conditioning system and a learning device that perform air conditioning using airflow.

[0002] The air conditioning system disclosed in Patent Document 1 aims to achieve automatic operation control that matches the user's operating preferences by learning a learning model based on operation history information of the indoor units.

[0003] Japanese Patent Application Laid-Open No. 2021-63611

[0004] In the air conditioning system disclosed in Patent Document 1, if two or more settings are possible for air conditioning operation, the settings made by the automatic operation control may not match the user's operating preferences. For example, some users prefer to be exposed to the airflow of the air conditioner, while others do not. If the system blows air toward a user who does not prefer to be exposed to the airflow, or blows air away from a user who prefers to be exposed to the airflow, the air conditioning operation may not match the user's preferences, even if the settings such as air volume match the user's operating preferences. In the air conditioning system disclosed in Patent Document 1, if the automatic operation control is performed in a way that does not match the user's operating preferences even though the current state matches the user's operating preferences, the user must manually change the settings made by the automatic operation control, which can be time-consuming to perform air conditioning control that matches the user's operating preferences.

[0005] The present disclosure has been made in consideration of the above, and aims to provide an air conditioning system that allows a user to decide whether or not to perform automatic driving control that matches the user's operating preferences.

[0006] In order to solve the above-mentioned problems and achieve the object, the air conditioning system according to the present disclosure includes an air conditioning appliance having an air conditioning unit that performs air conditioning operation and a control unit that controls the air conditioning unit, an inference data acquisition unit that acquires inference data including the air conditioning environment, which is information about the environment in which the air conditioning appliance is installed, an inference device that infers proposed settings for the air conditioning appliance from the air conditioning environment, including at least two settings that match a user's operating preferences for the air conditioning appliance, and a communication terminal that remotely operates the air conditioning appliance. The communication terminal includes a display unit that displays multiple icons representing the functions of the air conditioning appliance. The control unit displays proposed setting icons on the display unit for setting the proposed settings inferred by the inference device, and when an operation to press a proposed setting icon is performed, the control unit controls the air conditioning appliance according to the proposed setting corresponding to the proposed setting icon.

[0007] The air conditioning system according to the present disclosure has the effect of allowing the user to decide whether or not to perform automatic driving control that matches the user's operation preferences.

[0008] FIG. 1 is a diagram showing the configuration of an air conditioning system according to embodiment 1; functional block diagram of an air conditioning system according to embodiment 1; FIG. 1 is a diagram showing the configuration of a learning device for an air conditioning system according to embodiment 1; FIG. 1 is a diagram showing an example of learning data for an air conditioning system according to embodiment 1; FIG. 1 is a diagram showing the configuration of an inference device for an air conditioning system according to embodiment 1; FIG. 1 is a diagram showing an example of proposed settings for an air conditioning system according to embodiment 1; FIG. 1 is a diagram showing an example of a neural network used in a learning device for an air conditioning system according to embodiment 1;

[0009] An air conditioning system and a learning device according to an embodiment will be described in detail below with reference to the drawings.

[0010] Embodiment 1. Fig. 1 is a diagram showing the configuration of an air conditioning system according to Embodiment 1. The air conditioning system 1 includes air conditioning equipment 10, a communication terminal 20, and a server 30. The air conditioning equipment 10, the communication terminal 20, and the server 30 are connected to a network 40, and are capable of transmitting and receiving information to and from each other via the network 40. The air conditioning system 1 is capable of remotely controlling the operation of the air conditioning equipment 10 using the communication terminal 20. The communication terminal 20 is, for example, a smartphone terminal, a tablet terminal, or a laptop computer terminal. The network 40 is, for example, the Internet, but may also be a local network.

[0011] FIG. 2 is a functional block diagram of the air conditioning system according to the first embodiment. The air conditioning equipment 10 includes an air conditioning unit 13 that performs air conditioning operation, a control unit 11 that controls the air conditioning unit 13, and a remote controller 12 that serves as an interface for accepting user operations. The air conditioning unit 13 allows for the setting of multiple items for air conditioning control. Examples of setting items for the air conditioning control of the air conditioning unit 13 include a setting mode, air direction, air volume, air deflection, and functions used. Note that "air deflection" is a control that blows air toward occupants in an indoor space; when "air deflection" is on, air is blown toward the occupants, and when "air deflection" is off, air is blown away from the occupants. The control unit 11 controls the air conditioning unit 13 to perform air conditioning operation, thereby conditioning the indoor space.

[0012] When a user operates the remote controller 12 or the communication terminal 20, commands are input from the control unit 11 to the air conditioning unit 13. The commands include, for example, switching between on and off of the air conditioning operation, setting the operation mode, setting the temperature, setting the humidity, setting the air volume, setting the air direction, and setting the timer. The air conditioning unit 13 performs air conditioning operation in accordance with the input commands. Examples of operation modes include cooling operation, heating operation, and dehumidification operation.

[0013] The server 30 includes a learning device 34 that learns the relationship between at least two settings that match the user's operating preferences for air conditioning control of the air conditioning unit 13 and the ambient environment of the air conditioning equipment 10, a learned model memory unit 36 ​​that stores the learned model generated by the learning device 34, and an inference device 35 that uses the learned model to infer proposed settings 39 including at least two settings of the air conditioning equipment 10 from inference data 38.

[0014] FIG. 3 is a diagram illustrating the configuration of a learning device for an air conditioning system according to embodiment 1. The learning device 34 includes a learning data acquisition unit 341 that acquires learning data 37 and a model generation unit 342 that generates a trained model based on the learning data 37 acquired by the learning data acquisition unit 341. The learning data 37 includes history information 33, which is information about past settings of setting items for air conditioning control by the air conditioning unit 13, and past air conditioning environments 31. Note that the at least two past settings by the user for air conditioning control of the air conditioning unit 13 indicated by the history information 33 may include not only an operation for simultaneously setting multiple items, but also an operation for changing the setting of one setting item at intervals. An example of simultaneously setting multiple items is setting the temperature and airflow at the same time. An example of changing the setting of one setting item at intervals is changing the set airflow after a certain period of time has elapsed.

[0015] 4 is a diagram showing an example of learning data for the air conditioning system according to Embodiment 1. The history information 33 includes the details of a series of operations performed by a user to set the operation mode, set temperature, airflow direction, airflow volume, and the like via operations on the remote controller 12 or the communication terminal 20. The history information 33 may be stored in the air conditioning equipment 10 or in a database provided separately from the air conditioning equipment 10. The air conditioning environment 31 includes information such as the indoor temperature detected by a temperature sensor, the humidity detected by a humidity sensor, the floor temperature detected by an infrared sensor, the outdoor temperature detected by an outdoor air temperature sensor, the weather, and the amount of precipitation. The temperature sensor, humidity sensor, infrared sensor, and outdoor air temperature sensor may be provided in the air conditioning equipment 10 or may be provided separately from the air conditioning equipment 10.

[0016] The learning data acquisition unit 341 acquires the air conditioning environment 31 and the history information 33 as learning data 37. The learning data acquisition unit 341 can acquire the air conditioning environment 31 and the history information 33 as learning data 37 periodically, for example, once a day.

[0017] The model generation unit 342 uses the learning data 37 to learn the settings that the user regularly makes for the air conditioning equipment 10, and generates a trained model for inferring proposed settings 39 that include settings for at least two of the air conditioning equipment 10. The trained model generated by the learning device 34 is stored in the trained model storage unit 36. The trained model stored in the trained model storage unit 36 ​​is a model for inferring the proposed settings 39 from the air conditioning environment 31.

[0018] FIG. 5 is a diagram illustrating the configuration of an inference device for an air conditioning system according to embodiment 1. The inference device 35 includes an inference data acquisition unit 351 that acquires inference data 38 and an inference unit 352 that infers a proposed setting 39 from the inference data 38. The inference data 38 is the air conditioning environment 31. Therefore, the inference data acquisition unit 351 acquires the air conditioning environment 31. The inference unit 352 infers the proposed setting 39 from the air conditioning environment 31 acquired by the inference data acquisition unit 351 using a trained model stored in the trained model storage unit 36. That is, the inference unit 352 finds settings that are regularly performed by the user based on the air conditioning environment 31 acquired by the inference data acquisition unit 351 and the trained model acquired from the trained model storage unit 36. That is, by inputting the air conditioning environment 31 included in the inference data 38 acquired by the inference data acquisition unit 351 into this trained model, the proposed setting 39 inferred from the air conditioning environment 31 can be output.

[0019] 6 is a diagram showing an example of suggested settings for the air conditioning system according to Embodiment 1. The suggested settings 39, which are output data, are operation details for operating the air conditioning system with settings corrected based on the air conditioning environment 31, such as the indoor temperature, floor temperature, and outdoor temperature, and include two or more settings. For example, the suggested settings 39 may be an operation such as operating the system for 60 minutes with cooling at 20°C, upward airflow direction, strong airflow, and airflow on, followed by 120 minutes with cooling at 25°C, automatic airflow direction, weak airflow, and airflow off. In this way, the settings indicated by the suggested settings 39 may include operations for changing the settings of one or more items at time intervals.

[0020] Here, we have explained that the proposed settings 39 are output using a trained model generated by the model generation unit 342 of the air conditioning system 1, but it is also possible to obtain a trained model from outside, such as another air conditioning system, and output the proposed settings 39 based on this trained model.

[0021] The learning device 34 and the inference device 35 are used to learn the proposed settings 39 of the air conditioning unit 13, but may be, for example, devices separate from the air conditioning system 1 and connected to the air conditioning unit 13 via a network 40. Furthermore, the learning device 34 and the inference device 35 may exist on a cloud server.

[0022] The learning algorithm used by the model generation unit 342 may be a known algorithm such as supervised learning, unsupervised learning, reinforcement learning, etc. As an example, the case of supervised learning using a neural network will be described.

[0023] The model generation unit 342 learns the proposed settings 39 by so-called supervised learning, for example, according to a neural network model. Here, supervised learning refers to a method of providing the learning device 34 with data pairs of inputs and labels as results, thereby learning features in the learning data 37 and inferring results from the inputs.

[0024] 7 is a diagram showing an example of a neural network used in the learning device of the air conditioning system according to Embodiment 1. The neural network 70 is composed of an input layer 71 consisting of a plurality of neurons X1, X2, and X3, an intermediate layer 72 consisting of a plurality of neurons Y1 and Y2, and an output layer 73 consisting of a plurality of neurons Z1, Z2, and Z3. The intermediate layer may be one layer, or two or more layers. The intermediate layer 72 is also called a hidden layer.

[0025] For example, in a three-layer neural network 70 as shown in Figure 7, when multiple inputs are input to the input layer 71, the values ​​are multiplied by weights w11, w12, w13, w14, w15, and w16 before being input to the intermediate layer 72, and the results are further multiplied by weights w21, w22, w23, w24, w25, and w26 before being output from the output layer 73. This output result varies depending on the values ​​of the weights w11, w12, w13, w14, w15, and w16 and the values ​​of the weights w21, w22, w23, w24, w25, and w26.

[0026] The neural network 70 learns the proposed settings 39 by so-called supervised learning in accordance with the learning data 37, which is a combination of the air conditioning environment 31 and the history information 33 acquired by the learning data acquisition unit 341.

[0027] That is, the neural network 70 learns by inputting the air conditioning environment 31 into the input layer 71 and adjusting the weights w11, w12, w13, w14, w15, and w16 and the weights w21, w22, w23, w24, w25, and w26 so that the result output from the output layer 73 approaches the proposed setting 39.

[0028] The model generation unit 342 generates and outputs a trained model by performing the above-described learning.

[0029] The trained model storage unit 36 ​​stores the trained model output from the model generation unit 342.

[0030] FIG. 8 is a flowchart showing the procedure of the learning process of the air conditioning system according to the first embodiment.

[0031] In step S11, the learning data acquisition unit 341 acquires the air conditioning environment 31 and the history information 33, which are the learning data 37. Note that although the learning data acquisition unit 341 is described as acquiring the air conditioning environment 31 and the history information 33 simultaneously, it is sufficient that the learning data acquisition unit 341 acquires the air conditioning environment 31 and the history information 33 in association with each other, and the learning data acquisition unit 341 may acquire the data of the air conditioning environment 31 and the history information 33 at different times.

[0032] In step S12, the model generation unit 342 learns the proposed settings 39 by so-called supervised learning according to the learning data 37, which is a combination of the air conditioning environment 31 and historical information 33 acquired by the learning data acquisition unit 341, and generates a learned model.

[0033] In step S13, the trained model storage unit 36 ​​stores the trained model generated by the model generation unit 342.

[0034] Note that multiple regression analysis can also be applied to the learning algorithm used by the learning device 34. When multiple regression analysis is applied to the learning algorithm, the learning device 34 calculates a regression equation representing the air conditioning equipment 10, which is the target variable, using, for example, the air conditioning environment 31 and history information 33 of the air conditioning equipment 10 as explanatory variables. Also, deep learning, which learns to extract feature quantities themselves, can also be used as the learning algorithm used by the learning device 34. Furthermore, the learning device 34 may perform machine learning according to other known methods such as genetic programming, functional logic programming, support vector machines, etc.

[0035] The learning device 34 may generate a trained model for each user based on the air conditioning environment 31 and the user's history information 33. By using the trained model for each user, the inference device 35 can infer suggested settings 39 that are in line with the settings that the target user regularly makes.

[0036] FIG. 9 is a flowchart showing the procedure of the inference process of the air conditioning system according to the first embodiment.

[0037] In step S21, the inference data acquisition unit 351 acquires the air conditioning environment 31, which is the inference data 38.

[0038] In step S22, the inference unit 352 inputs the air conditioning environment 31 into the learned model stored in the learned model storage unit 36 ​​and obtains the proposed setting 39.

[0039] In step S23, the inference unit 352 outputs the proposed settings 39 obtained by the learned model to the air conditioning equipment 10.

[0040] In step S24, the control unit 11 causes the display unit 15 of the communication terminal 20 to display, in addition to the icons of the existing functions, an operation icon 190 which is a proposed setting icon corresponding to the proposed setting 39. The user can execute the proposed setting 39 obtained in step S13 from the operation icon 190 newly displayed on the display unit 15.

[0041] 10 is a diagram showing an example of an icon display area displayed on the display unit of the communication terminal of the air conditioning system according to Embodiment 1. The communication terminal 20 has an icon display area 16 in a part of the display unit 15.

[0042] 11 is an enlarged view of an example of an icon display area displayed on the display unit of the communication terminal of the air conditioning system according to Embodiment 1. In the icon display area 16, the order in which operation icons are displayed is fixed. Operation icons 180, 170, and 160 for executing functions 1, 2, and 3 are already displayed and are not shortcut buttons. An operation icon 190 corresponding to a newly displayed proposed setting 39 is automatically arranged in the icon display area 16. When the proposed setting 39 is output from the inference unit 352, a new icon is displayed in the icon display area 16.

[0043] In FIG. 11 , the operation icon 180 for function 1 is in the first display position. The operation icon 170 for function 2 is in the second display position. The operation icon 160 for function 3 is in the third display position. The operation icon 190 for suggested settings is in the fourth display position. The operation icons 180, 170, and 160 for functions 1, 2, and 3 are existing icons, but the operation icon 190 for suggested settings is newly displayed through periodic data acquisition and learning, such as once a day. By pressing the operation icon 190, the control unit 11 performs air conditioning control in accordance with the suggested settings 39. Note that if some of the settings included in the suggested settings 39 have already been set by the user, the operation icon 190 will be used to additionally set items that have not been set by the user.

[0044] Note that, by pressing the operation icon 190, details of the operation of the proposed settings 39 may be displayed on the display unit 15. FIG. 12 is a diagram showing an example of an air conditioning setting screen of the air conditioning system according to embodiment 1. In the example shown in FIG. 12, operation details such as operating the air conditioner at 18°C, with an upward airflow direction and an air volume of 4 for 30 minutes, followed by operating the air conditioner at 25°C, with an automatic airflow direction and an air volume of 2 for 3 hours are displayed on the display unit 15. When an execute button 191 displayed on the display unit 15 along with the operation details is pressed, the control unit 11 performs air conditioning control in accordance with the proposed settings 39. In this way, the user can confirm the contents of the air conditioning settings according to the proposed settings 39 and then have the control unit 11 perform air conditioning control in accordance with the proposed settings 39.

[0045] The air conditioning system 1 according to the first embodiment is equipped with an inference device 35 that infers proposed settings 39 from the air conditioning environment 31, which include at least two settings that match the user's operating preferences for the air conditioning equipment 10. By pressing the proposed setting icon 190 corresponding to the proposed setting 39, air conditioning control is performed in accordance with the proposed setting 39, and the user can therefore decide whether or not to perform automatic operation control that matches the user's operating preferences.

[0046] The following describes the hardware configuration of the learning device 34 and the inference device 35 according to Embodiment 1. Fig. 13 is a diagram showing the hardware configuration of the learning device of the air conditioning system according to Embodiment 1. The learning device 34 is realized by a computer system including a processor 91 that executes various processes, a memory 92 that is a main memory, and a storage device 93 that stores information.

[0047] The processor 91 may be a computing means such as an arithmetic unit, a microprocessor, a microcomputer, a CPU (Central Processing Unit), or a DSP (Digital Signal Processor). The memory 92 may be a non-volatile or volatile semiconductor memory such as a RAM (Random Access Memory), a ROM (Read Only Memory), a flash memory, an EPROM (Erasable Programmable Read Only Memory), or an EEPROM (Electrically Erasable Programmable Read Only Memory). The storage device 93 stores a program for performing a process of learning the proposed settings 39.

[0048] In the computer system described above, the processor 91 reads into the memory 92 a program stored in the storage device 93 and corresponding to the processing of each component, and executes the program, thereby realizing the functions of the model generation unit 342. The memory 92 is also used as a temporary memory for each process executed by the processor 91. The program executed by the processor 91 may be provided in a state stored in a storage medium, or may be provided via the network 40.

[0049] Like the learning device 34, the inference device 35 is realized by a computer system including a processor 91 that executes various processes, a memory 92 that serves as a main memory, and a storage device 93 that stores information. The processor 91 reads out programs corresponding to the processes of each component stored in the storage device 93 into the memory 92 and executes them, thereby realizing the functions of the inference unit 352. The memory 92 is also used as a temporary memory for each process executed by the processor 91. The programs executed by the processor 91 may be provided in a state where they are stored in a storage medium, or may be provided via the network 40.

[0050] The configurations shown in the above embodiments are merely examples of the content, and may be combined with other known technologies, or parts of the configurations may be omitted or modified without departing from the spirit of the invention.

[0051] 1 Air conditioning system, 10 Air conditioning equipment, 11 Control unit, 12 Remote controller, 13 Air conditioning unit, 15 Display unit, 16 Icon display area, 20 Communication terminal, 30 Server, 31 Air conditioning environment, 33 History information, 34 Learning device, 35 Inference device, 36 Learned model storage unit, 37 Learning data, 38 Inference data, 39 Proposed settings, 40 Network, 70 Neural network, 71 Input layer, 72 Intermediate layer, 73 Output layer, 91 Processor, 92 Memory, 93 Storage device, 160, 170, 180, 190 Operation icon, 191 Execution button, 341 Learning data acquisition unit, 342 Model generation unit, 351 Inference data acquisition unit, 352 Inference unit.

Claims

1. An air conditioning system comprising: an air conditioning appliance having an air conditioning unit that performs air conditioning operation and a control unit that controls the air conditioning unit; an inference data acquisition unit that acquires inference data including the air conditioning environment, which is information about the environment in which the air conditioning appliance is installed; an inference device that infers from the air conditioning environment proposed settings including at least two settings that match a user's operating preferences for the air conditioning appliance; and a communications terminal that remotely operates the air conditioning appliance, wherein the communications terminal has a display unit that displays a plurality of icons representing functions of the air conditioning appliance, the control unit causes the display unit to display proposed setting icons for making the proposed settings inferred by the inference device, and when an operation is performed to press one of the proposed setting icons, the control unit controls the air conditioning appliance in accordance with the proposed setting corresponding to the proposed setting icon.

2. An air conditioning system as described in claim 1, wherein the inference device infers the proposed settings based on the air conditioning environment input from the inference data acquisition unit using a learned model for inferring the proposed settings from the air conditioning environment.

3. An air conditioning system according to claim 1 or 2, wherein the air-conditioned environment is information on at least one of an outdoor temperature and a floor temperature.

4. An air conditioning system according to any one of claims 1 to 3, wherein the proposed settings are settings consisting of at least two of indoor temperature, wind direction, air volume and wind deflection.

5. An air conditioning system as described in claim 4, wherein the proposed settings include two or more settings that are executed with an execution interval therebetween.

6. An air conditioning system according to any one of claims 1 to 5, further comprising a server equipped with the inference data acquisition unit and the inference device, wherein the air conditioning equipment and the server are connected via a network.

7. An air conditioning system as described in any one of claims 1 to 6, comprising a learning device having a learning data acquisition unit that acquires learning data including historical information including at least two settings by a user for the air conditioning environment and the air conditioning equipment, and a model generation unit that uses the learning data to generate a trained model for inferring proposed settings including at least two settings by the user for the air conditioning equipment from the air conditioning environment.

8. An air conditioning system according to claim 7, wherein the history information is information indicating settings for at least two of an operation mode, a set temperature, an airflow direction, an airflow rate, and an air blower.

9. A learning device comprising: a learning data acquisition unit that acquires learning data including an air conditioning environment, which is information about the environment in which an air conditioning appliance is installed, and historical information including at least two settings by a user for the air conditioning appliance; and a model generation unit that uses the learning data to generate a trained model for inferring proposed settings, including at least two settings by a user for the air conditioning appliance, from the air conditioning environment.

10. The learning device according to claim 9, wherein the history information is information indicating settings for at least two of an operation mode, a set temperature, an airflow direction, an airflow rate, and an air blowing direction.

Images