Air conditioning system, adjustment device, adjustment method and program

Abstract

To adjust air conditioning control according to a user's taste.SOLUTION: An air conditioning system 1 includes: an air conditioner 20 which includes an air conditioning part 203 for air-conditioning an internal space, and an air conditioning control part 202 for controlling the air conditioning part 203; a remote controller 30 for operating the air conditioner 20; a history storage part 103 for storing history of an instruction command which is the instruction command based on a user's operation with respect to the remote controller 30 and which is expressed in a natural language into a storage part 104, as a command history; a taste estimation part 105 for analyzing the command history stored in the storage part 104 by the time zone, and for estimating a user's taste; and an adjustment part 106 for adjusting the control by the air conditioning control part 202 based on the user's taste estimated by the taste estimation part 105.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present disclosure relates to an air conditioning system, an adjustment device, an adjustment method, and a program. [Background technology]

[0002] In the field of air conditioners, the temperature and air volume of the discharged air are adjusted using control techniques such as PID (Proportional-Integral-Differential) control to condition the space to be air-conditioned.

[0003] As an example of a control technology, Patent Document 1 discloses a technology in which, in PID control, a special operation is performed in which the manipulated variable is oscillated at a constant amplitude, and control parameters are adjusted based on the control variable obtained during the special operation. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Application Laid-Open No. 2009-151521 Summary of the Invention [Problem to be solved by the invention]

[0005] When a user changes the settings of an air conditioner using a remote controller (hereinafter referred to as "remote control"), the remote control is operated according to the user's preferences. Examples of user preferences include frequently changing the set temperature, changing settings to emphasize energy conservation, etc.

[0006] If the control of the air conditioner could be adjusted in accordance with the user's preferences, it would be possible to perform air conditioning control that is more suited to the user's preferences.

[0007] In view of the above circumstances, an object of the present disclosure is to provide an air conditioning system or the like that is capable of adjusting air conditioning control in accordance with user preferences. [Means for solving the problem]

[0008] In order to achieve the above object, the air conditioning system according to the present disclosure comprises: an air conditioner including an air conditioning unit that conditions an indoor space and an air conditioning control unit that controls the air conditioning unit; a remote controller for operating the air conditioner; a history storage means for storing a history of command commands expressed in natural language based on user operations on the remote controller as a command history in a storage means; preference estimation means for estimating the user's preferences by analyzing the command history stored in the storage means for each time period; an adjustment unit that adjusts the control by the air conditioning control unit based on the user's preference estimated by the preference estimation unit; Equipped with. [Effects of the Invention]

[0009] According to the present disclosure, air conditioning control can be adjusted according to the user's preferences. [Brief explanation of the drawings]

[0010] [Figure 1] FIG. 1 is a diagram showing an overall configuration of an air conditioning system according to a first embodiment of the present disclosure. [Figure 2] FIG. 10 is a diagram illustrating an example of a command conversion table used in the adjustment device according to the first embodiment of the present disclosure. [Figure 3] FIG. 1 is a diagram illustrating an example of a command history stored in an adjustment device according to the first embodiment of the present disclosure. [Figure 4] FIG. 1 is a diagram illustrating an example of a rule-based model used in an adjustment device according to a first embodiment of the present disclosure. [Figure 5] FIG. 10 is a diagram showing an example of adjustment content of air conditioning control according to preferences by an adjustment device according to the first embodiment of the present disclosure. [Figure 6] FIG. 1 is a diagram illustrating an example of fuzzy control in an air conditioning system according to the first embodiment of the present disclosure. [Figure 7] FIG. 1 is a diagram illustrating an example of a hardware configuration of an adjustment device according to a first embodiment of the present disclosure. [Figure 8] 1 is a flowchart illustrating an example of an operation of saving a command history by an adjustment device according to a first embodiment of the present disclosure. [Figure 9] 1 is a flowchart showing an example of an operation for adjusting air conditioning control by an adjustment device according to the first embodiment of the present disclosure. [Figure 10] FIG. 10 is a diagram showing the overall configuration of an air conditioning system according to a fourth embodiment of the present disclosure. [Figure 11] FIG. 10 is a diagram showing the overall configuration of an air conditioning system according to a fourth embodiment of the present disclosure. DETAILED DESCRIPTION OF THE INVENTION

[0011] Hereinafter, an air conditioning system according to an embodiment of the present disclosure will be described with reference to the drawings. In each drawing, the same or equivalent parts are denoted by the same reference numerals.

[0012] (Embodiment 1) An air conditioning system 1 according to a first embodiment will be described with reference to FIG. 1. The air conditioning system 1 is an air conditioning system that conditions the indoor space of a user's home. The air conditioning system 1 includes an adjustment device 10, an air conditioner 20, and a remote control 30. As will be described in detail below, the adjustment device 10 estimates a user's preferences based on the user's operation of the remote control 30, and adjusts the air conditioning control of the air conditioner 20 based on the estimated preferences. For example, when the estimated user preference is to prioritize energy conservation, the adjustment device 10 adjusts the control parameters of the air conditioning control of the air conditioner 20 so as to be suitable for energy conservation. As a result, the air conditioning system 1 can adjust the air conditioning control in accordance with the user's preferences. The air conditioning system 1 is an example of an air conditioning system according to the present disclosure.

[0013] As described above, the adjustment device 10 estimates the user's preferences and adjusts the air conditioning control of the air conditioner 20 based on the estimated preferences. The adjustment device 10 is communicatively connected to the air conditioner 20. The adjustment device 10 is communicatively connected to the air conditioner 20 by, for example, a dedicated communication line. The adjustment device 10 adjusts the air conditioning control of the air conditioner 20 via, for example, this communication line.

[0014] Similarly to the air conditioner 20 described below, the adjustment device 10 receives operation signals issued by a remote control 30. The dashed lines in FIG. 1 indicate the flow of operation signals issued by the remote control 30. As will be described in detail later, the adjustment device 10 estimates the user's preferences by analyzing a command history obtained based on operation signals issued by the remote control 30. The functional configuration of the user's adjustment device 10 will be described later. The adjustment device 10 is an example of an adjustment device according to the present disclosure.

[0015] The air conditioner 20 is an air conditioner that conditions the indoor space of a user's home. The air conditioner 20 is, for example, a room air conditioner equipped with an indoor unit and an outdoor unit. Like the adjustment device 10, the air conditioner 20 receives operation signals emitted by a remote control 30. The air conditioner 20 performs air conditioning in accordance with the operation signals received from the remote control 30. However, control-related settings such as control algorithms and control parameters used when controlling the air conditioning are adjusted by the adjustment device 10 based on the user's preferences. The functional configuration of the air conditioner will be described later. The air conditioner 20 is an example of an air conditioner according to the present disclosure.

[0016] The remote control 30 is a remote controller that allows a user to operate the air conditioner 20. The remote control 30 includes, for example, a plurality of buttons, and issues operation signals such as starting operation, ending operation, changing the operation mode, and changing the set temperature in response to button operations. The remote control 30 issues, for example, infrared signals. As described above, the operation signals issued by the remote control 30 are received by both the adjustment device 10 and the air conditioner 20. The remote control 30 is an example of a remote controller according to the present disclosure.

[0017] Continuing to refer to FIG. 1, the functional configuration of the adjustment device 10 and the air conditioner 20 will be described.

[0018] The adjustment device 10 includes a signal receiving unit 101, a command conversion unit 102, a history saving unit 103, a memory unit 104, a preference estimation unit 105, and an adjustment unit 106.

[0019] The signal receiving unit 101 receives an operation signal emitted by the remote control 30. The signal receiving unit 101 includes, for example, a receiving module that receives an infrared signal.

[0020] The command conversion unit 102 converts the operation signal received by the signal receiving unit 101 into a command expressed in a natural language. The natural language is, for example, Japanese, English, etc. In the following explanation, we will explain the case where Japanese is used as the natural language.

[0021] The command conversion unit 102 converts the operation content indicated by the operation signal into a command expressed in simple Japanese words, for example, based on the conversion table shown in Fig. 2. The command to be converted includes both "temperature + / -" and "set temperature X°C" because the operation to change the set temperature itself and the result of how many degrees Celsius the set temperature has become can each affect the estimation of the user's preference.

[0022] Referring again to Fig. 1, the history storage unit 103 stores the history of the instruction commands obtained by the command conversion unit 102 in the storage unit 104. Specifically, the history storage unit 103 associates the instruction commands with the dates and times when the instruction commands were obtained and stores them in the storage unit 104 as a command history. The command history stored in the storage unit 104 is, for example, as shown in Fig. 3. The history storage unit 103 is an example of a history storage means according to the present disclosure.

[0023] As described above, the storage unit 104 stores the command history. The storage unit 104 also stores a model, which will be described later. The storage unit 104 is an example of a storage means according to the present disclosure.

[0024] The preference estimation unit 105 uses the model stored in the storage unit 104 to analyze the command history stored in the storage unit 104 for each time period to estimate the user's preferences. The "time period" here refers to a time period separated by a fixed amount of time, such as a time period separated by one hour, such as 9:00-10:00 or 11:00-12:00, or a time period separated by 30 minutes. Alternatively, a time period when the user is likely to be awake, such as from 7:00 AM to 10:00 PM, may be separated into one-hour time periods, and a time period when the user is likely to be asleep, such as from 10:00 PM to 7:00 AM, may be separated into three-hour time periods. The preference estimation unit 105 is an example of a preference estimation means according to the present disclosure.

[0025] The preference estimation unit 105 estimates the user's preferences by analyzing the command history using a rule-based model. The rules set in the model are, for example, as shown in Fig. 4. According to the rules shown in Fig. 4, if the user frequently changed the temperature setting during the same time period on the previous day, it is estimated that the user's preference is to prioritize response time.

[0026] Although the rules shown in FIG. 4 all refer to command histories relating to the same time period on the previous day, they may also refer to other dates and times.

[0027] Referring again to Fig. 1, the adjustment unit 106 adjusts the air conditioning control by the air conditioning control unit 202 of the air conditioner 20, which will be described later, based on the user's preferences estimated by the preference estimation unit 105. The adjustment unit 106 determines the adjustment content from the estimated user's preferences, for example, according to the table shown in Fig. 5. The adjustment unit 106 applies the determined adjustment content to the air conditioning control unit 202 of the air conditioner 20, and adjusts the air conditioning control by the air conditioning control unit 202. The adjustment unit 106 is an example of an adjustment means according to the present disclosure.

[0028] As an example of adjustment, the following describes the cases where the control method for air conditioning control is PID control, fuzzy control, or model predictive control.

[0029] First, we will explain the case of PID control. In PID control, control is performed based on the following equation:

[0030]

number

[0031] In PID control, the response time can be shortened by increasing the proportional gain Kp. Also, if priority is given to control accuracy rather than response time, the control accuracy can be improved by decreasing the proportional gain Kp.

[0032] Next, we will explain the case of fuzzy control. In fuzzy control, a table map such as that shown in Fig. 6 is prepared. In fuzzy control, vague criteria such as "large," "normal," and "small" are expressed using membership functions, and the table map shown in Fig. 6 is referenced based on these criteria to control the air conditioning in accordance with each control law.

[0033] When the adjustment unit 106 adjusts the fuzzy control, it changes the table map itself, for example, shown in FIG. 6, in accordance with the estimated user preferences. For example, while both the air volume and temperature difference in FIG. 6 have three levels, a table map is created in which both have four levels. This improves control accuracy. However, because the control becomes more complex, the risk of control instability also increases.

[0034] Next, we will explain the case of model predictive control. Model predictive control is a control technique that uses a predictive model of the controlled object to predict future responses at each time and performs optimization. Here, we will describe the implementation method with reference to the literature (A. Bemporad, "Model Predictive Control Design: New Trends and Tools," Proceedings of the 45th IEEE Conference on Decision and Control, San Diego, CA, USA, 2006, pp. 6678-6683, doi: 10.1109 / CDC.2006.377490.).

[0035] In the following, k is the current discrete time, x(k) is the state of the controlled object, and u(k) is the control input to the controlled object. For example, if the controlled object is an air conditioner, x(k) is the room temperature and humidity, and u(k) is the compressor frequency and fan rotation speed of the air conditioner.

[0036] In model predictive control, first, the cost function J(k) is determined using the following formula (1). Here, P, Q, and R are adjustable weighting matrices. By tuning each element of this weighting matrix, the response performance and robustness of model predictive control can be adjusted. Each element of P, Q, and R is set according to the estimated user preferences. Note that the prime symbols attached to x(k) and u(k) in formula (1) indicate transposition.

[0037]

number

[0038] Then, under the constraints described below, an input sequence U of control inputs from time k to k+Np-1 is calculated using the following equation (2) to minimize the cost function J(k). Note that U is defined as U=u(k),u(k+1), ,u(k+Np-1), where Np is an integer called the prediction horizon.

[0039]

number

[0040] The constraints are given by the following equations (3) and (4). Equation (3) represents a prediction model of the controlled object, and is a model of the controlled object using a linear state space model. Equation (4) represents a constraint that defines the lower limit umin and upper limit umax of the control input u, thereby limiting the operating range of the control input.

[0041]

number

[0042]

number

[0043] Equation (1) is in the form of an equation to which the Quadratic Problem, a quadratic form mathematical optimization method, can be applied. Therefore, with the prediction model in equation (3) and the operating conditions of the control input in equation (4) as constraints, it is possible to calculate U that minimizes J(k) using equation (2).

[0044] When calculating equation (2), the constraint condition for the control input u in equation (4) can be applied to the control input quantities of the refrigeration cycle, such as the compressor frequency and fan rotation speed, to impose upper and lower limits. By using such constraint conditions, for example, it is possible to place an upper limit on the fan rotation speed and perform control that prioritizes quietness. In addition, by placing a constraint on the upper limit of the compressor frequency, which consumes a lot of power, it is possible to reduce power consumption.

[0045] Note that a system identification method can be used to create the predictive model of equation (3). For example, when the user is absent or there are no control requirements such as comfort or quietness, the control input of the room air conditioner can be varied, the room temperature and humidity can be collected, and the controlled object can be modeled using a system identification method such as the least squares method or the N4SID method. The state space model created in this way is used in the predictive model of model predictive control of equation (3).

[0046] 1, the functional configuration of the air conditioner 20 will be described. The air conditioner 20 comprises a signal receiving unit 201, an air conditioning control unit 202, and an air conditioning unit 203.

[0047] The signal receiving unit 201 receives an operation signal emitted by the remote control 30. The signal receiving unit 201 includes, for example, a receiving module that receives an infrared signal.

[0048] The air conditioning control unit 202 controls the air conditioning by the air conditioning unit 203 based on the operation signal received by the signal receiving unit 201. For example, when the operation signal indicates a change in the set temperature, the air conditioning control unit 202 controls the discharge temperature in accordance with the operation signal. However, as described above, the air conditioning control by the air conditioning control unit 202 is adjusted by the adjustment unit 106 of the adjustment device 10. The air conditioning control unit 202 is an example of an air conditioning control means according to the present disclosure.

[0049] The air conditioning unit 203 conditions the indoor space of the user's house under the control of the air conditioning control unit 202. The air conditioning unit 203 includes main units of an air conditioner, such as a compressor, louvers, a fan, and a heat exchanger. The air conditioning unit 203 is an example of an air conditioning means according to the present disclosure.

[0050] An example of the hardware configuration of the adjustment device 10 will be described with reference to Fig. 7. The adjustment device 10 shown in Fig. 7 is realized by a computer such as a personal computer or a microcontroller.

[0051] The adjustment device 10 includes a processor 1001, a memory 1002, an interface 1003, and a secondary storage device 1004, which are connected to each other via a bus 1000.

[0052] The processor 1001 is, for example, a CPU (Central Processing Unit). The processor 1001 loads an operating program stored in the secondary storage device 1004 into the memory 1002 and executes the program, thereby realizing each function of the adjustment device 10.

[0053] The memory 1002 is a main storage device configured, for example, by RAM (Random Access Memory). The memory 1002 stores the operating program that the processor 1001 reads from the secondary storage device 1004. The memory 1002 also functions as a working memory when the processor 1001 executes the operating program.

[0054] The interface 1003 is an I / O (Input / Output) interface such as a serial port, a USB (Universal Serial Bus) port, a network interface, etc. For example, by connecting a receiving module to the interface 1003, the function of the signal receiving unit 101 is realized.

[0055] The secondary storage device 1004 is, for example, a flash memory, a hard disk drive (HDD), or a solid state drive (SSD). The secondary storage device 1004 stores the operating program executed by the processor 1001. The secondary storage device 1004 also realizes the function of the storage unit 104.

[0056] An example of the operation of saving a command history by the adjustment device 10 will be described with reference to Fig. 8. The operation shown in Fig. 8 is performed constantly while the adjustment device 10 is in operation.

[0057] The signal receiving unit 101 of the adjusting device 10 waits for an operation signal from the remote controller 30 (step S101).

[0058] When the signal receiving unit 101 receives an operation signal from the remote control 30, the command converting unit 102 converts the received operation signal into a command expressed in a natural language (step S102).

[0059] The history saving unit 103 saves the instruction command obtained in step S102 as a command history in the storage unit 104 in association with the obtained date and time (step S103). Then, the adjustment device 10 repeats the operations from step S101.

[0060] An example of the operation of adjusting air conditioning control by the adjusting device 10 will be described with reference to Fig. 9. The operation shown in Fig. 9 is executed at regular intervals, such as every 30 minutes or every hour. Alternatively, it may be executed at the start of each "time period" defined by the preference estimation unit 105.

[0061] The preference estimation unit 105 of the adjustment device 10 analyzes the command history stored in the storage unit 104 for each time period (step S111).

[0062] The preference estimation unit 105 estimates the user's preferences based on the analysis results in step S111, using the rule-based model stored in the storage unit 104 (step S112).

[0063] The adjustment unit 106 determines the adjustment content based on the user's preferences estimated in step S112 (step S113).

[0064] The adjustment unit 106 adjusts the air conditioning control by the air conditioning control unit 202 of the air conditioner 20 in accordance with the adjustment content determined in step S113 (step S114). Then, the adjusting device 10 ends the operation of adjusting the air conditioning control.

[0065] The above describes the air conditioning system 1 according to Embodiment 1. In the air conditioning system 1, the adjustment device 10 converts operation signals issued by a user operating the remote control 30 into command commands expressed in natural language and saves them as a command history, analyzes the command history using a model to estimate the user's preferences, and adjusts air conditioning control based on the estimated preferences. In this way, the air conditioning system 1 can adjust air conditioning control according to the user's preferences.

[0066] (Modification of the first embodiment) In the first embodiment, the adjustment device 10 is a device separate from the air conditioner 20, but the air conditioner 20 may have the adjustment device 10 built in. In this case, the signal receiving unit 101 of the adjustment device 10 and the signal receiving unit 201 of the air conditioner 20 may be the same. In this case, the air conditioner 20 is an example of the adjustment device 10 according to the present disclosure.

[0067] In the first embodiment, the adjustment device 10 includes the storage unit 104, but the storage unit 104 may be replaced with an external storage device. For example, the adjustment device 10 may have a communication function via the Internet, and a storage server on the Internet may be used instead of the storage unit 104. In this case, the storage server is an example of a storage means according to the present disclosure.

[0068] (Embodiment 2) In the first embodiment, a rule-based model is used to estimate a user's preferences. In the second embodiment, a model for performing latent semantic analysis is used instead to estimate a user's preferences. The overall configuration of the air conditioning system 1 in the second embodiment is the same as that in the first embodiment shown in FIG. 1 , and is the same as the first embodiment except that the model used by the preference estimation unit 105 is different. Therefore, only the differences in the models will be described below.

[0069] A typical model for latent semantic analysis is the topic model. Another example of latent semantic analysis using a topic model is LDA (Latent Dirichlet Analysis). LDA makes it possible to classify documents by inferring latent topics within them.

[0070] A topic model is a machine learning model for classifying documents by regarding each document as consisting of multiple topics. Each topic is composed of a set of words. The types of words and their frequency of occurrence vary depending on the topic.

[0071] As mentioned above, command history is expressed in natural language, so by treating the set of command history contained in a time period as a document and analyzing it using LDA with a topic model, the command history for each time period can be classified by topic.

[0072] For example, if a user's preference for cooling comfort is high during a certain time period, the remote control 30 will be operated based on the preference for energy conservation during that time period, and it can be expected that the command history for that time period will be implicitly classified as a topic that prioritizes energy conservation. However, because LDA estimates and classifies latent topics, it is not possible to directly grasp what kind of topic it is at the time of classification.

[0073] It should be noted that, since it is preferable to have a certain amount of documents when classifying using a topic model, it is preferable to store command history for about 10 days or more.

[0074] Then, focusing on one time period, the topic to which the most command histories belong for each day is regarded as the topic for that time period. Then, by analyzing the words that frequently appear in the commands related to that topic, it is possible to estimate the user's preferences for that time period.

[0075] For example, let's say we focus on the time period between 9:00 and 10:00. Suppose that only the command history from 9:00 to 10:00 three days ago in the past week is classified as Topic A, and the command history from 9:00 to 10:00 on other days is classified as Topic B. In this case, Topic B is the most frequently classified command during the time period between 9:00 and 10:00. Therefore, the topic for the time period between 9:00 and 10:00 is considered to be Topic B. Then, by analyzing the words that appear frequently in the command history classified as Topic B, we can estimate the user's preferences for the time period between 9:00 and 10:00.

[0076] If a user's preferences are estimated simply by referring to the command history for the same time period on the previous day, as in embodiment 1, an unusual remote control operation on the previous day will affect the estimation of the user's preferences for the next day. On the other hand, in embodiment 2, latent semantic analysis is used to classify the command history for each time period over multiple days by topic and estimate preferences for each time period, thereby reducing the impact of unusual remote control operations.

[0077] Latent semantic analysis can also be performed using a Dynamic Topic Model (DTM). By using a DTM, it is possible to classify and exclude characteristics of specific days of the week, change the model depending on how the air conditioner 20 is used as the seasons change, and so on.

[0078] (Embodiment 3) In the first embodiment, a rule-based model is used to estimate a user's preferences. In the third embodiment, instead, a natural language processing model constructed using a neural network is used to estimate a user's preferences. The overall configuration of the air conditioning system 1 in the third embodiment is the same as that in the first embodiment shown in FIG. 1 , and is the same as the first embodiment except for the difference in the model used by the preference estimation unit 105. Therefore, only the difference in the model will be described below.

[0079] As a natural language processing model constructed using a neural network, for example, a large language model (LLM) such as Word2Vec, BERT, Transformer, or GPT can be used.

[0080] By inputting each command contained in the command history for each time period into the LLM and examining the word vectors output by the LLM, the command history can be represented as vector data embeddings in the document space (Word Embeddings).

[0081] For example, the manufacturer of the adjustment device 10 can build a model for classifying new command histories into clusters by collecting command histories from a large number of users in advance and inputting them into the LLM. Then, using the LLM built in this way, it is possible to estimate user preferences from the degree of conformance of the new command history to each cluster.

[0082] (Fourth embodiment) An air conditioning system 1 according to embodiment 4 will be described with reference to Figure 10. The air conditioning system 1 according to embodiment 4 differs from embodiment 1 in that it further includes an electrical device 40 and a relay server. It also differs from embodiment 1 in that the adjustment device 10 further includes a communication unit 107. It also differs from embodiment 1 in that the command conversion unit 102 has an additional function, as will be described later. Note that the functional configuration of the air conditioner 20 is the same as in embodiment 1, and therefore the description of the air conditioner 20 in Figure 10 is simplified.

[0083] Electrical appliance 40 is an electrical appliance operated by a user, and is, for example, a home appliance installed in a user's home. Although only one electrical appliance 40 is shown in FIG. 10, there may be multiple electrical appliances 40. Electrical appliance 40 is communicably connected to relay server 50. When electrical appliance 40 is operated by a user, it outputs a signal indicating the user operation to relay server 50. Electrical appliance 40 is an example of an electrical appliance according to the present disclosure.

[0084] The relay server 50 is communicatively connected to the electric device 40 and the adjustment device 10. The relay server 50 acquires a signal indicating a user operation output by the electric device 40, and accumulates an equipment operation history that is a history of the user's operations on the electric device 40. The relay server 50 outputs the equipment operation history to the adjustment device 10.

[0085] The communication unit 107 of the adjustment device 10 receives the device operation history output by the relay server 50. The communication unit 107 includes, for example, a network interface.

[0086] In addition to the operations in the first embodiment, the command conversion unit 102 of the adjustment device 10 in the fourth embodiment converts the device operation history into instruction commands. As a result, the command history stored in the history storage unit 103 includes instruction commands related to the operation of the electric device 40.

[0087] As a result, the preference estimation unit 105 takes into account the user's operation of the electrical appliance 40. For example, when the history of the user operating the cooking heater is included, the user's preference can be estimated as "emphasis on cooking," and the adjustment unit 106 can adjust the control in anticipation of, for example, fluctuations in the heat load in the indoor space.

[0088] (Embodiment 5) In the first embodiment, the user's preferences at the present time are estimated, but future user preferences may also be estimated. For example, when it is currently 9:00, the user's preferences for 10:00-11:00 may be estimated, and air conditioning control may be adjusted in advance to satisfy those preferences.

[0089] (Sixth embodiment) An air conditioning system 1 according to embodiment 6 will be described with reference to Figure 11. The air conditioning system 1 according to embodiment 6 differs from embodiment 1 in that it further includes a HEMS controller 60 connected to the power grid network NT. As with embodiment 4, it also differs from embodiment 1 in that the adjusting device 10 further includes a communication unit 107 and that the command conversion unit 102 has additional functions. Note that the functional configuration of the air conditioner 20 is the same as in embodiment 1, and therefore the illustration of the air conditioner 20 in Figure 11 is simplified.

[0090] The HEMS controller 60 is a HEMS controller that controls HEMS (Home Energy Management System) devices. The HEMS controller 60 is connected to HEMS-compatible devices (not shown). Although not shown, the air conditioner 20 is also a HEMS-compatible device and is connected to the HEMS controller 60.

[0091] The HEMS controller 60 receives a demand-response control signal from the power grid network NT and performs demand-response control on each HEMS-compatible device. By demand-response control, for example, when the balance between the amount of power on the demand side and the amount of power on the supply side in the power grid network NT is about to be lost, the HEMS-compatible devices in the home are controlled from outside to stabilize the power grid network NT.

[0092] The HEMS controller 60 outputs the information indicated by the demand response control signal to the adjustment device 10.

[0093] The communication unit 107 of the adjustment device 10 receives information indicated by the demand response control signal from the HEMS controller 60.

[0094] In addition to the operations of the first embodiment, the command conversion unit 102 of the adjustment device 10 in the sixth embodiment converts information indicated by the demand-response control signal into a command. As a result, the command history stored in the history storage unit 103 includes command commands related to demand-response control.

[0095] As a result, demand-response control is also taken into consideration in the preference estimation by the preference estimation unit 105. For example, air conditioning control can be adjusted on the assumption that the air conditioner 20 will be stopped during time periods when demand-response control is likely to occur.

[0096] (Embodiment 7) In the first embodiment, an environmental sensor may be further provided, the environmental sensor may be communicably connected to the adjustment device 10, and information about the environment detected by the environmental sensor may be converted into an instruction command and saved as a command history.

[0097] For example, a temperature and humidity sensor may be installed outside the house, and the temperature and humidity outside detected by the temperature and humidity sensor may be converted into command commands and used as a command history. Also, an air quality sensor may be installed inside the house, and information on the carbon dioxide concentration, fine particulate matter concentration, etc. detected by the air quality sensor may be converted into command commands and used as a command history. Also, a temperature and humidity sensor installed in the outdoor unit may be used as an environmental sensor.

[0098] This allows the environment inside and outside the home to be reflected in the estimation of the user's preferences.

[0099] (Embodiment 8) In the first embodiment, the user may wear a biometric sensor, and the information output by the biometric sensor may be converted into command instructions and used as a command history. For example, the user may wear a device including a biometric sensor, such as a smart watch or a pulse monitor, and the information output by the device may be converted into command instructions and used as a command history.

[0100] This allows the user's state to be reflected in the estimation of the user's preferences.

[0101] (Other variations)

[0102] 7, the adjustment device 10 includes a secondary storage device 1004. However, the present invention is not limited to this, and the secondary storage device 1004 may be provided outside the adjustment device 10, and the adjustment device 10 and the secondary storage device 1004 may be connected via an interface 1003. In this configuration, removable media such as a USB flash drive or a memory card may also be used as the secondary storage device 1004.

[0103] 7, the adjustment device 10 may be configured by a dedicated circuit using an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array), etc. In the hardware configuration shown in FIG. 7, some of the functions of the adjustment device 10 may be realized by a dedicated circuit connected to the interface 1003, for example.

[0104] Various aspects of the present disclosure are summarized below as appendices.

[0105] (Appendix 1) an air conditioner including an air conditioning unit that conditions an indoor space and an air conditioning control unit that controls the air conditioning unit; a remote controller for operating the air conditioner; a history storage means for storing a history of command commands expressed in natural language based on user operations on the remote controller as a command history in a storage means; preference estimation means for estimating the user's preferences by analyzing the command history stored in the storage means for each time period; an adjustment unit that adjusts the control by the air conditioning control unit based on the user's preference estimated by the preference estimation unit; An air conditioning system equipped with: (Appendix 2) the preference estimation means estimates the user's preference by classifying the command using a model for performing latent semantic analysis in analyzing the command history; 1. An air conditioning system as described in Appendix 1. (Appendix 3) the preference estimation means estimates the user's preference by classifying the command using a model for performing natural language processing in analyzing the command history. 1. An air conditioning system as described in Appendix 1. (Appendix 4) The history storage means further stores in the storage means a history of operations by the user on electrical appliances installed in the user's home as the command history. 10. An air conditioning system according to any one of claims 1 to 3. (Appendix 5) The preference estimation means further estimates future preferences of the user by analyzing the command history; The adjustment means further adjusts the control by the air conditioning control means based on the future preferences of the user estimated by the preference estimation means. 10. An air conditioning system according to any one of claims 1 to 4. (Appendix 6) The history storage means further stores, as the command history, a history of demand response control transmitted from an external power network of the user's home. 10. An air conditioning system according to any one of claims 1 to 5. (Appendix 7) the history storage means further stores in the storage means a history of information about the environment detected by the environmental sensor as the command history; 10. An air conditioning system according to any one of claims 1 to 6. (Appendix 8) The history storage means further stores a history of the user's biological signals in the storage means as the command history. 8. An air conditioning system according to any one of claims 1 to 7. (Appendix 9) a history storage means for storing, as a command history, a history of command commands expressed in natural language based on user operations on a remote controller for operating the air conditioner; and preference estimation means for estimating the user's preferences by analyzing the command history stored in the storage means for each time period; an adjustment means for adjusting control of the air conditioner based on the user's preference estimated by the preference estimation means; An adjustment device comprising: (Appendix 10) storing a history of command commands expressed in natural language based on user operations on a remote controller for operating the air conditioner as a command history in a storage means; analyzing the command history stored in the storage means for each time period to estimate the user's preferences; adjusting control of the air conditioner based on the estimated preferences of the user; Adjustment method. (Appendix 11) Computer, a history storage means for storing, as a command history, a history of command commands expressed in natural language based on user operations on a remote controller for operating the air conditioner; preference estimation means for estimating the preferences of the user by analyzing the command history stored in the storage means for each time period; an adjustment means for adjusting control of the air conditioner based on the user's preference estimated by the preference estimation means; A program that functions as a [Explanation of symbols]

[0106] 1 Air conditioning system, 10 Adjustment device, 20 Air conditioner, 30 Remote control, 40 Electrical equipment, 50 Relay server, 60 HEMS controller, 101 Signal receiving unit, 102 Command conversion unit, 103 History storage unit, 104 Memory unit, 105 Preference estimation unit, 106 Adjustment unit, 107 Communication unit, 1000 Bus, 1001 Processor, 1002 Memory, 1003 Interface, 1004 Secondary storage device, NT Power grid network.

Claims

1. an air conditioner including an air conditioning unit that conditions an indoor space and an air conditioning control unit that controls the air conditioning unit; a remote controller for operating the air conditioner; a history storage means for storing a history of command commands expressed in natural language based on user operations on the remote controller as a command history in a storage means; preference estimation means for estimating the user's preferences by analyzing the command history stored in the storage means for each time period; an adjustment unit that adjusts the control by the air conditioning control unit based on the user's preference estimated by the preference estimation unit; An air conditioning system equipped with:

2. the preference estimation means estimates the user's preference by classifying the command using a model for performing latent semantic analysis in analyzing the command history; The air conditioning system of claim 1 .

3. the preference estimation means estimates the user's preference by classifying the command using a model for performing natural language processing in analyzing the command history. The air conditioning system of claim 1 .

4. The history storage means further stores in the storage means a history of operations by the user on electrical appliances installed in the user's home as the command history. An air conditioning system according to any one of claims 1 to 3.

5. The preference estimation means further estimates future preferences of the user by analyzing the command history; The adjustment means further adjusts the control by the air conditioning control means based on the future preferences of the user estimated by the preference estimation means. The air conditioning system of claim 1 .

6. The history storage means further stores, as the command history, a history of demand response control transmitted from an external power network of the user's home. The air conditioning system of claim 1 .

7. the history storage means further stores in the storage means a history of information about the environment detected by the environmental sensor as the command history; The air conditioning system of claim 1 .

8. The history storage means further stores a history of the user's biological signals in the storage means as the command history. The air conditioning system of claim 1 .

9. a history storage means for storing, as a command history, a history of command commands expressed in natural language based on user operations on a remote controller for operating the air conditioner; and preference estimation means for estimating the user's preferences by analyzing the command history stored in the storage means for each time period; an adjustment means for adjusting control of the air conditioner based on the user's preference estimated by the preference estimation means; An adjustment device comprising:

10. storing a history of command commands expressed in natural language based on user operations on a remote controller for operating the air conditioner as a command history in a storage means; analyzing the command history stored in the storage means for each time period to estimate the user's preferences; adjusting control of the air conditioner based on the estimated preferences of the user; Adjustment method.

11. Computer, a history storage means for storing, as a command history, a history of command commands expressed in natural language based on user operations on a remote controller for operating the air conditioner; preference estimation means for estimating the preferences of the user by analyzing the command history stored in the storage means for each time period; an adjustment means for adjusting control of the air conditioner based on the user's preference estimated by the preference estimation means; A program that functions as a

Images