Proposal device and proposal system

Abstract

A proposal device (100) is provided with a computation device (101), a storage device (102), an input interface (103), and an output interface (104). The input interface (103) receives building information and an index type. The storage device (102) stores first determination data (110) and second determination data (120). The computation device (101) obtains specification candidates by using the first determination data (110) and obtains evaluations of the specification candidates by using the second determination data (120). The output interface (104) outputs display information to a display device (20), the display information being information for displaying the specification candidates and the evaluations of the specification candidates to a user.

Description

Proposed device and proposed system

[0001] The present disclosure relates to a proposal device and a proposal system, and more particularly to a proposal device and a proposal system that proposes specifications of an air conditioner to a user.

[0002] Because there are many different specifications for air conditioners, selecting an air conditioner is not an easy task for users considering installing one. For this reason, various support systems have been proposed to assist users in the selection process.

[0003] For example, Japanese Patent Laid-Open Publication No. 2011-123617 (Patent Document 1) describes a support system that takes into consideration the layout of a room and the insulation specifications of a building and assists in selecting an air conditioner that achieves high energy-saving performance.

[0004] JP 2011-123617 A

[0005] However, users do not always select air conditioners with only an emphasis on energy-saving performance. With environmental destruction caused by global warming becoming a concern, some users may want to select an air conditioner using CO2 emissions as an indicator, for example. Users who are in a hurry to install an air conditioner will likely want to select an air conditioner using construction time as an indicator. In such cases, the magnitude of the evaluation based on the indicators may vary depending on the conditions of the building in which the air conditioner is installed.

[0006] The present disclosure aims to provide a technology that can suggest appropriate air conditioner specifications to a user that take into account the building's conditions, based on indicators that correspond to the user's intentions.

[0007] The present disclosure provides a proposal device that proposes air conditioner specifications to a user, the proposal device comprising a calculation device, a storage device, an input interface that accepts information input from a user, and an output interface that outputs information, wherein the input interface accepts building information regarding a building in which the air conditioner is to be installed and an index type for evaluating the performance of the air conditioner, the storage device is configured to store first judgment data used to obtain candidate specifications for the air conditioner and second judgment data used to obtain an evaluation of the candidate specifications, the calculation device uses the first judgment data to obtain candidate specifications suitable for the building information and uses the second judgment data to obtain an evaluation of the candidate specifications based on the index type, the evaluation of the candidate specifications being an evaluation based on the performance that the candidate specification air conditioner is estimated to exhibit if it is installed in a building identified by the building information, and the output interface outputs display information to a display device to allow a user to identify the candidate specifications and the evaluation of the candidate specifications.

[0008] Another aspect of the present disclosure is a proposal system that proposes air conditioner specifications to a user, comprising: a communication device; a computing device configured to communicate with the communication device via a network; and a storage device configured to store first judgment data used to acquire candidate specifications for the air conditioner and second judgment data used to acquire an evaluation of the candidate specifications; when the communication device receives building information about a building in which the air conditioner is to be installed and an index type for evaluating the performance of the air conditioner, the communication device transmits the building information and the index type to the computing device; the computing device uses the first judgment data to acquire candidate specifications suitable for the building information and uses the second judgment data to acquire an evaluation of the candidate specifications based on the index type; and transmits display information to the communication device to allow a user to identify the candidate specifications and the evaluation of the candidate specifications; the evaluation of the candidate specifications is based on the performance that the candidate air conditioner is estimated to exhibit if it is installed in the building identified by the building information; and the communication device displays the display information on a display device.

[0009] According to the present disclosure, it is possible to provide a technology that can propose appropriate air conditioner specifications to a user that take into account the building conditions, based on an index according to the user's intentions.

[0010] 1 is a block diagram illustrating the configuration of a proposed device according to a first embodiment. FIG. 2 is a diagram illustrating an example of the configuration of an air conditioner and various information related to the selection of an air conditioner. FIG. 3 is a block diagram illustrating machine learning of a first inference model. FIG. 4 is a block diagram illustrating machine learning of a second inference model. FIG. 5 is a diagram illustrating specific examples of building information and specific examples of air conditioner specifications. FIG. 6 is a diagram illustrating specific examples of index types. FIG. 7 is a block diagram illustrating the processing of the proposed device (when the first inference model and the second inference model MA are used). FIG. 8 is a block diagram illustrating the processing of the proposed device (when the first inference model and the second inference model MB are used). FIG. 9 is a flowchart illustrating the processing of the proposed device. FIG. 10 is a diagram illustrating an example of a screen displayed on the display of a display device based on display information output from the proposed device. FIG. 11 is a flowchart illustrating the processing of a proposed device according to a modified example. FIG. 11 is a block diagram illustrating the configuration of a proposed device according to a second embodiment. FIG. 12 is a block diagram illustrating the processing of the proposed device (when the first database and the second inference model MA are used). FIG. 13 is a block diagram illustrating the configuration of a proposed system according to a fourth embodiment.

[0011] Each embodiment will be described in detail below with reference to the drawings. The same or corresponding parts in the drawings are designated by the same reference numerals, and their description will not be repeated. The size relationships of the components in the following drawings may differ from the actual ones. It is anticipated that the embodiments and modifications in this disclosure may be combined with each other.

[0012] Embodiment 1. [Proposal Device 100] Fig. 1 is a block diagram showing the configuration of a proposal device 100 according to embodiment 1. The proposal device 100 has a function of proposing candidate specifications for an air conditioner to a user, taking into consideration information about the building in which the air conditioner is installed. The proposal device 100 is realized, for example, by a server device or a personal computer. The proposal device 100 includes a calculation device 101, a storage device 102, an input interface 103, and an output interface 104.

[0013] The arithmetic device 101 is a computing entity (computer) that executes various processes according to various programs. The arithmetic device 101 includes a processor, which is an example of a control circuit. The arithmetic device 101 includes, for example, at least one of a central processing unit (CPU), a field programmable gate array (FPGA), a graphics processing unit (GPU), and a multi-processing unit (MPU). Furthermore, the arithmetic device 101 may include volatile memory such as dynamic random access memory (DRAM) and static random access memory (SRAM), and non-volatile memory such as read-only memory (ROM) and flash memory. The arithmetic device 101 may be configured with a processing circuitry.

[0014] The storage device 102 includes non-volatile memory such as a hard disk drive (HDD) and a solid state drive (SSD). The storage device 102 stores various programs and data. The storage device 102 stores at least a first inference model 110, a second inference model 120, and an inference program 130. The second inference model 120 includes a second inference model MA, a second inference model MB, a second inference model MC, etc. Each of the first inference model 110 and the second inference model 120 is a model trained by machine learning. The arithmetic device 101 performs inference processing using the first inference model 110 and the second inference model 120. The arithmetic device 101 performs inference processing according to the inference program 130. The arithmetic device 101 may perform various processes using dedicated hardware (electronic circuits) rather than software.

[0015] The input interface 103 is connected to the input device 10. The input device 10 is composed of, for example, a keyboard, a mouse, a microphone, etc. The output interface 104 is connected to the display device 20 having a display 21.

[0016] Before describing the operation of the proposed device 100, the background that led to the proposal of the proposed device 100 in this disclosure will be described.

[0017] [Background Description] Fig. 2 is a diagram illustrating an example of the configuration of an air conditioner and various information related to the selection of an air conditioner. As shown in Fig. 2, an air conditioner 50 includes an outdoor unit 51 and an indoor unit 52.

[0018] The outdoor unit 51 includes a compressor 1, an outdoor heat exchanger 2, a four-way valve 5, and a fan 6 provided corresponding to the outdoor heat exchanger 2. The indoor unit 52 includes an indoor heat exchanger 3, an expansion valve 4, and a fan 7 provided corresponding to the indoor heat exchanger 3. The outdoor unit 51 and the indoor unit 52 include a refrigerant circuit capable of circulating refrigerant. The refrigerant circuit includes the compressor 1, the outdoor heat exchanger 2, the indoor heat exchanger 3, the expansion valve 4, the four-way valve 5, and refrigerant piping 8.

[0019] 2 merely illustrates an air conditioner 50 having a basic configuration. There are a wide variety of air conditioners on the market with different specifications depending on the air conditioning method, capacity range, number of indoor units, and options (e.g., salt damage resistance specifications). While the variety of air conditioner types meets the various needs of users, it also complicates the task of users selecting an air conditioner.

[0020] As shown in Figure 2, a user selects the specifications of an air conditioner while taking into consideration the installation location of the air conditioner. At this time, the user may select the specifications of the air conditioner using power consumption as an indicator, but the indicator used by the user in selecting the specifications of the air conditioner is not necessarily limited to power consumption. For example, a user who wants to replace a broken air conditioner will likely want to select the specifications of the air conditioner by prioritizing the construction period over power consumption. With environmental destruction caused by global warming becoming a concern, some users may want to select the specifications of the air conditioner using CO2 emissions as an indicator.

[0021] As such, the types of indicators that are considered important when selecting air conditioner specifications may differ depending on the user. Therefore, various types of indicators should be offered to users selecting air conditioner specifications according to their needs. However, air conditioner specifications based on indicators should not be evaluated based on general test results, etc., listed in catalogs, etc.

[0022] This is because the index value is thought to differ depending on the location of the building in which the user installs the air conditioner (for example, cold region, warm region, etc.), the age and use of the building, the layout of the building, etc. Therefore, when evaluating the specifications of an air conditioner based on the index selected by the user, information about the building in which the user installs the air conditioner must be taken into consideration.

[0023] Therefore, this disclosure describes a suggestion device that can suggest to a user appropriate air conditioner specifications that take into account the building's conditions, based on an index that corresponds to the user's intentions.

[0024] A proposing device 100 according to the first embodiment has the configuration shown in Fig. 1. The proposing device 100 proposes candidate specifications for an air conditioner to a user. Hereinafter, the "candidate specifications for an air conditioner" may be simply referred to as the "candidate specifications."

[0025] The user operates the input device 10 to input building information. The user further operates the input device 10 to input an index type of interest. The user operating the input device 10 may be a person who desires the installation of an air conditioner, or may be a person who operates the input device 10 while interacting with the person who desires the installation of an air conditioner.

[0026] Building information is information about a building in which air conditioners are planned to be installed. Building information may be, for example, one or more of the floor plan, location, age, and purpose shown in FIG. 2. Use refers to the use of the building. Specific examples of building uses include hotels, offices, stores, hospitals, etc. Index types are, for example, the amount of power consumption, CO2 emissions, installation costs, construction period, etc. shown in FIG. 2. Here, in order to distinguish between each index type, the multiple index types are referred to as index type A, index type B, index type C, etc.

[0027] [Input Interface 103 and Output Interface 104] The input interface 103 shown in Figure 1 accepts building information input from a user. The input interface 103 also accepts an index type input from a user. The computing device 101 uses a first inference model 110 to infer candidate specifications of air conditioners suitable for the building. Furthermore, the computing device 101 uses a second inference model 120 to infer the magnitude of the evaluation of the candidate specifications based on the index type.

[0028] 1 outputs display information to the display device 20 to allow the user to identify the specification candidates and their evaluations. The display device 20 displays the specification candidates for air conditioners on the display 21 together with evaluations based on index types specified by the user. The user refers to the evaluations to select specification candidates that interest them.

[0029] [Machine Learning of First Inference Model 110] Figure 3 is a block diagram for explaining machine learning of the first inference model 110. As shown in Figure 3, the first inference model 110 is trained by supervised learning in a training unit 205 of the computer 201. The training unit 205 is composed of hardware and software of the computer 201. The training unit 205 accepts building information and specification candidates as learning data to be used in the machine learning of the first inference model 110.

[0030] Of the building information and the specification candidates, the specification candidates are the correct answer data. Here, the "specification candidates" are specification candidates for air conditioners that are considered desirable to install in a building identified by the building information. The training unit 205 accepts a large amount of data set consisting of the building information and specification candidates and trains the first inference model 110. The trained first inference model 110 is stored in the storage device 102 of the proposal device 100.

[0031] In this way, the first inference model 110 is trained to infer candidate specifications suitable for the building information based on learning data including the building information and candidate specifications. The first inference model 110 is an example of first determination data used to acquire candidate specifications for an air conditioner.

[0032] [Machine Learning of Second Inference Model 120] Figure 4 is a block diagram for explaining machine learning of the second inference model 120. As shown in Figure 4, the second inference model 120 is trained by supervised learning in a training unit 305 of a computer 301. The training unit 305 is configured by the hardware and software of the computer 301. Note that the second inference model 120 may be trained in the training unit 205 of the computer 201, similar to the first inference model 110.

[0033] The training unit 305 receives building information, specification candidates, and index values ​​as learning data used in machine learning of the second inference model 120. Of the building information, specification candidates, and index values, the index values ​​are the correct answer data. The training unit 205 receives a large amount of data set consisting of building information, specification candidates, and index values, and trains the second inference model 120.

[0034] 4, the second inference model 120 includes a second inference model MA, a second inference model MB, a second inference model MC, ... The second inference model MA corresponds to indicator type A, the second inference model MB corresponds to indicator type B, and the second inference model MC corresponds to indicator type C. The second inference model MA is trained based on the indicator values ​​of indicator type A. The second inference model MB is trained based on the indicator values ​​of indicator type B. The second inference model MC is trained based on the indicator values ​​of indicator type C.

[0035] For example, if indicator type A is power consumption, the second inference model MA is trained based on the amount of power consumption corresponding to the index value. If indicator type B is CO2 emissions, the second inference model MB is trained based on the CO2 emissions corresponding to the index value. If indicator type C is implementation cost, the second inference model MC is trained based on the implementation cost corresponding to the index value. The trained second inference model 120 is stored in the memory device 102 of the proposal device 100.

[0036] In this way, the second inference model 120 is trained to infer index values ​​corresponding to index types based on learning data including building information, specification candidates, and index values ​​corresponding to index types. The second inference model 120 is an example of second judgment data used to obtain an evaluation of the specification candidates. The index type A is an example of a first index, and the index type B is an example of a second index. The second inference model MA is an example of a first index model corresponding to the first index, and the second inference model MB is an example of a second index model corresponding to the second index.

[0037] [Specific Examples of Building Information, Air Conditioner Specifications, and Index Types] Fig. 5 is a diagram showing specific examples of building information and air conditioner specifications, and Fig. 6 is a diagram showing specific examples of index types.

[0038] As shown in Figure 5, building information may be composed of, for example, four pieces of data: location, age, purpose, and floor plan. "Location" refers to the location of the building where the air conditioner will be installed. "Age" refers to the age of the building where the air conditioner will be installed. "Use" refers to the purpose of the building where the air conditioner will be installed (hotel, office, store, etc.). "Floor plan" refers to the floor plan of the building where the air conditioner will be installed, and more specifically, refers to the floor plan of the room where the indoor unit will be placed. "Floor plan" includes the shape and dimensions of the room. "Floor plan" includes the ceiling height. "Floor plan" may also include the floor plan of the space where the outdoor unit is installed.

[0039] Other examples of building information (a1) to (a7) are listed below: Inspection and repair history (whether or not inspections and repairs were performed, the locations inspected and repaired, the dates and details of the inspections and repairs), (a1) Building height, (a2) Weather information (including outside air speed), (a3) ​​Number of people in the room, (a4) Door open / close information, (a5) Lighting device usage information (ON / OFF), (a6) Blind open / close information, and (a7) Window orientation. As shown in FIG. 5 , the specifications of an air conditioner may be composed of, for example, two pieces of data: (air conditioning method, capacity range). The "air conditioning method" includes the "individual distribution method" and the "centralized duct method." The "individual distribution method" refers to a method in which multiple indoor units, each equipped with an air outlet and an air inlet, are arranged at intervals. The "centralized duct method" refers to a method in which conditioned air from a single indoor unit is blown out from multiple air outlets via an air duct. "Capacity band" means the air conditioner capacity (rated capacity kW) per unit floor area.

[0040] Other examples of air conditioner specifications (b1) to (b9) are listed below: (b1) Air conditioner options (salt-resistant specifications, etc.) (b2) Indoor unit installation location (b3) Outdoor unit installation location (including distance from adjacent outdoor units, including height from ground level) (b4) Number of indoor units (b5) Intake port orientation of indoor unit (b6) Outlet orientation of indoor unit (b7) Outlet orientation of outdoor unit (b8) Intake port orientation of outdoor unit (b9) Distance to other outdoor units As shown in FIG. 6 , indicator types A, B, C, etc. may be configured to include the power consumption of the air conditioner, CO2 emissions from manufacture to disposal of the air conditioner, COP (Coefficient of Performance) of the air conditioner, installation cost of the air conditioner, and construction period (number of days for installation work) of the air conditioner.

[0041] Instead of "power consumption," "running costs," "electricity charges," or "primary energy consumption" may be used. "COP" may be the average COP over a certain period. "Installation costs" are referred to as initial costs and include the cost of equipment and construction costs.

[0042] Other examples of index types (c1) to (c6) are listed below: (c1) Inspection time or number of days for an air conditioner (strictly speaking, inspection time per unit time, for example, inspection time per year) (c2) Number of inspection points for an air conditioner (for example, the sum of the number of indoor units and the number of outdoor units) (c3) Repair time or number of days for an air conditioner (strictly speaking, inspection time per unit time, for example, repair time per 10 years) (c4) Repair cost for an air conditioner (strictly speaking, inspection time per unit time, for example, total repair cost per 10 years) (c5) Lifespan of an air conditioner (number of years from purchase to replacement) (c6) Thermal comfort of occupants (PMV (Predicted Mean Vote), etc.; strictly speaking, average thermal comfort of occupants) [Processing of the Proposal Device 100 Based on an Inference Model] FIGS. 7 and 8 are block diagrams for explaining the processing of the proposal device 100. Figure 7 shows the case where the first inference model 110 and the second inference model MA are used, and Figure 8 shows the case where the first inference model 110 and the second inference model MB are used.

[0043] First, a case in which the first inference model 110 and the second inference model MA are used will be described with reference to Figure 7. When the user specifies index type A among multiple index types, the proposal device 100 infers the index value of index type A using the second inference model MA.

[0044] As shown in Figure 7, building information and index type (=A) are input to the proposal device 100. The user inputs the building information and index type to the proposal device 100 using the input interface 103 (see Figure 1). The calculation device 101 included in the proposal device 100 determines to use the second inference model MA corresponding to index type A from among the multiple second inference models 120.

[0045] The calculation device 101 inputs building information to the first inference model 110. The first inference model 110 outputs candidate specifications for an air conditioner that are suitable for the building information as an inference result. In this way, by inputting the building information to the first inference model 110, the calculation device 101 obtains candidate specifications that are suitable for the building information from the first inference model 110.

[0046] The specification candidate output as the inference result may be one or more. Here, it is assumed that the first inference model 110 outputs three specification candidates (specification candidate 1, specification candidate 2, and specification candidate 3) in response to the input of building information. In other words, in the present disclosure, it is assumed that there are multiple optimal solutions for the first inference model 110. For example, the global optimal solution and all local optimal solutions may correspond to the optimal solutions of the first inference model 110. In this way, by inputting building information into the first inference model 110, the computing device 101 obtains air conditioner specification candidates 1 to 3 that are suitable for the building information from the first inference model 110.

[0047] The calculation device 101 inputs building information and specification candidate 1 obtained from the first inference model 110 to the second inference model MA. The second inference model MA outputs an index value of index type A corresponding to specification candidate 1 and the building information as an inference result.

[0048] The calculation device 101 inputs building information and specification candidate 2 obtained from the first inference model 110 to the second inference model MA. The second inference model MA outputs an index value of index type A corresponding to specification candidate 2 and the building information as an inference result.

[0049] The calculation device 101 inputs building information and specification candidate 3 obtained from the first inference model 110 to the second inference model MA. The second inference model MA outputs an index value of index type A corresponding to specification candidate 3 and the building information as an inference result.

[0050] In this way, the calculation device 101 outputs inference results for each specification candidate from the second inference model MA. The calculation device 101 acquires each inference result. Although three second inference models MA are depicted in FIG. 7, this shows the inference operation of the second inference model MA for each specification candidate. Therefore, the three second inference models MA depicted in FIG. 7 do not mean that the proposed device 100 is equipped with three second inference models MA.

[0051] In this way, by inputting building information and specification candidates into the second inference model MA, the calculation device 101 obtains an evaluation of the specification candidate based on index type A, i.e., an index value of index type A, from the second inference model 120. The index value of index type A output from the second inference model MA is a value estimated to be derived when the specification candidate air conditioner is installed in a building based on the building information. This value is significantly more reliable than index values ​​derived based on theoretical values, etc., listed in an air conditioner catalog, etc., because it takes into account the building in which the air conditioner will actually be installed. An evaluation based on such index values ​​is based on the performance estimated to be exhibited by the specification candidate air conditioner 50 when it is installed in a building identified by the building information.

[0052] Therefore, the user can select the air conditioner specifications that best meet the user's needs from among multiple specification candidates based on the specification candidates output from the first inference model 110 and the index values ​​output from the second inference model MA.

[0053] Next, a case in which the first inference model 110 and the second inference model MB are used will be described with reference to Figure 8. When the user specifies index type B among multiple index types, the proposal device 100 infers the index value of index type B using the second inference model MB.

[0054] 8, building information and an index type (=B) are input to the proposal device 100. The calculation device 101 determines to use the second inference model MB corresponding to the index type B from among the multiple second inference models 120.

[0055] The calculation device 101 then acquires specification candidates 1 to 3 from the first inference model 110 and acquires index values ​​from the second inference model MB. The index values ​​acquired from the second inference model MB are "the index value of index type B corresponding to specification candidate 1 and building information," "the index value of index type B corresponding to specification candidate 2 and building information," and "the index value of index type B corresponding to specification candidate 3 and building information."

[0056] The information input to the first inference model 110 has already been explained using Figure 7. The information input to the second inference model MB is the same as the information input to the second inference model MA. Since the information input to the second inference model MA has already been explained using Figure 7, that explanation will not be repeated here.

[0057] The above describes the processing of the proposal device 100 when the user specifies index type A and index type B, using Figures 7 and 8. When the user specifies any of index types C, D, etc., the proposal device 100 executes processing similar to that shown in Figures 7 and 8 using the first inference model 110 and any of the second inference models MC, MD, etc. corresponding to the index type. As a result, the proposal device 100 outputs specification candidates 1 to 3 suitable for the building information input by the user, and index values ​​corresponding to the building information and each specification candidate. The index values ​​output from the proposal device 100 are index values ​​of the index type specified by the user.

[0058] When the index type received by the input interface 103 is index type A, the calculation device uses a second inference model MA, which is an example of a first index model, to obtain an evaluation of the specification candidate based on the index type, and when the index type received by the input interface 103 is index type B, the calculation device uses a second inference model MB, which is an example of a second index model, to obtain an evaluation of the specification candidate based on the index type.

[0059] Fig. 9 is a flowchart for explaining the processing of the proposal device 100. Figs. 10 and 11 are diagrams showing examples of screens displayed on the display 21 of the display device 20 based on display information output from the proposal device 100. Here, the processing of the proposal device 100 will be explained in more detail according to the flowchart shown in Fig. 9. Note that the screens displayed on the display 21 will also be explained with reference to Fig. 10 or Fig. 11 as necessary.

[0060] First, the arithmetic device 101 accepts building information input from the user (step S101). As shown in (1) of FIG. 10, a screen prompting the user to input building information is displayed on the display 21. The user inputs the location, age, purpose, and layout of the building in which the air conditioner will be installed. After inputting the building information, the user clicks the OK button. This causes the arithmetic device 101 to accept the input of the building information via the input interface 103. Note that, to reduce the user's input burden, the screen may be configured to display options for input items in a pull-down format.

[0061] Next, the arithmetic device 101 inputs building information into the first inference model 110 (step S102). The first inference model 110 infers candidate air conditioner specifications suitable for the input building information and outputs the inference results. Next, the arithmetic device 101 acquires candidate air conditioner specifications from the first inference model 110 (step S103). For example, the arithmetic device 101 acquires candidate air conditioner specifications 1 to 3 from the first inference model 110. In this way, the arithmetic device 101 uses the first inference model 110 to acquire candidate air conditioner specifications suitable for the building information.

[0062] Next, the arithmetic device 101 displays the candidate specifications for the air conditioner (step S104). More specifically, the arithmetic device 101 outputs display information for displaying candidate specifications 1 to 3 for the air conditioner to the display device 20 via the output interface 104. As a result, candidate specifications 1 to 3 are displayed on the display 21, as shown in (2) of FIG.

[0063] The user checks the air conditioning method and capacity range of each of the specification candidates 1 to 3 displayed on the screen, and then clicks the OK button. As a result, the screen on the display 21 switches from (2) in Fig. 10 to (3) in Fig. 11. As shown in (3) in Fig. 11, a screen prompting the user to specify an index type is displayed on the display 21. After specifying the index type of interest, the user clicks the OK button.

[0064] The calculation device 101 receives an input of an index type via the input interface 103 (step S105). Next, the calculation device 101 sets a second inference model 120 corresponding to the input index type (step S106). For example, when the calculation device 101 receives index type A, it sets the second inference model MA, and when the calculation device 101 receives index type B, it sets the second inference model MB.

[0065] Next, the arithmetic device 101 sequentially inputs the candidate specifications 1 to 3 of the air conditioner into the set second inference model 120 (step S107). Next, the arithmetic device 101 acquires index values ​​corresponding to the candidate specifications 1 to 3 of the air conditioner from the second inference model 120 (step S108).

[0066] In this way, the calculation device 101 obtains index values ​​of the specification candidates based on the index type using the second inference model 120. Next, the calculation device 101 determines the magnitude of the evaluation based on the obtained index values ​​(step S109).

[0067] For example, if the acquired index value is an index type corresponding to power consumption, the larger the index value, the higher the electricity bill. Therefore, if the index value is an index type corresponding to power consumption, the rating of a specification candidate corresponding to a smaller index value should be higher than the rating of a specification candidate corresponding to a larger index value. From this perspective, the arithmetic device 101 determines the rating of a specification candidate according to the index type and the index value. The arithmetic device 101 may express the rating as a number of "stars" or as a numerical value. Here, an example in which the rating is expressed as a number of "stars" will be described.

[0068] The arithmetic device 101 displays each of the air conditioner specification candidates 1 to 3 in association with an evaluation based on the index type (step S110). More specifically, the arithmetic device 101 outputs display information for displaying each of the air conditioner specification candidates 1 to 3 in association with an evaluation based on the index type to the display device 20 via the output interface 104.

[0069] As a result, as shown in (4) of Fig. 11, specification candidates 1 to 3 and the evaluation of each of specification candidates 1 to 3 are displayed on the display 21. The index type designated by the user is also displayed on the display 21. The user refers to the evaluation and selects the specification candidate to be introduced into the building from specification candidates 1 to 3.

[0070] [Modification] Fig. 12 is a flowchart for explaining the processing of the proposed device according to a modification. In the flowchart shown in Fig. 9, the timing at which the calculation device 101 accepts input of building information (step S101) and the timing at which the calculation device 101 accepts input of index types (step S105) are different from each other.

[0071] However, as shown in step S201 of Fig. 12, the calculation device 101 may also accept both the building information and the index type. In this case, step S104 (a step of displaying candidate air conditioners) and step S105 (a step of accepting input of the index type) shown in Fig. 9 are unnecessary in the flowchart related to the modified example.

[0072] The flowchart relating to the modified example is the same as the flowchart shown in Fig. 9 except that step S201 is used instead of step S101 and steps S104 and S105 are deleted. Therefore, the description of the flowchart will not be repeated here.

[0073] As described above, according to the first embodiment, a user can select an air conditioner by taking into consideration the performance that the air conditioner is likely to exhibit in a building in which the user is considering installing the air conditioner. Moreover, the user can select an air conditioner by taking into consideration evaluations based on indexes of index types that the user is most interested in. As a result, according to the first embodiment, it is possible to propose to the user appropriate air conditioner specifications that take into consideration the building's conditions, based on indexes that correspond to the user's preferences.

[0074] Second Embodiment Next, a second embodiment will be described with reference to Fig. 13 and Fig. 14. Fig. 13 is a block diagram showing the configuration of a proposed device 100A according to the second embodiment. Fig. 14 is a block diagram for explaining the processing of the proposed device 100A (when a first database 110DB and a second inference model MA are used).

[0075] 13, the proposal device 100A according to the second embodiment differs from the proposal device 100 according to the first embodiment in that a first database 110DB is used instead of the first inference model 110, and that a selection program 131 is used to select necessary information from the first database 110DB. In other respects, the configuration of the proposal device 100A is the same as that of the proposal device 100.

[0076] As shown in Fig. 14, the first database 110DB has registered therein a table 111 that associates building information with specification candidates. In the table 111, three specification candidates 1 to 3 are associated with one piece of building information. Here, the "specification candidates" registered in the table 111 are "specification candidates suitable for the building information." As already explained, the "specification candidates suitable for the building information" are specification candidates for air conditioners that are considered desirable to install in a building identified by the building information.

[0077] When the calculation device 101 receives input of building information, it refers to the table 111 of the first database 110. The calculation device 101 searches the table 111 for specification candidates 1 to 3 that are registered in correspondence with the received building information.

[0078] The arithmetic device 101 retrieves specification candidates 1 to 3 that are found as a result of the search from the first database 110DB. The arithmetic device 101 inputs the retrieved specification candidates 1 to 3 into the second inference model 120 in the same manner as in the first embodiment. Thereafter, the arithmetic device 101 retrieves index values ​​corresponding to each of the specification candidates 1 to 3 from the second inference model 120 in the same manner as in the first embodiment. Note that while FIG. 14 shows a case in which the second inference model MA is used, the arithmetic device 101 executes processing in the same manner using the second inference models MA, MB, MC, etc., depending on the index type specified by the user.

[0079] According to embodiment 2, compared to embodiment 1, a first database 110DB capable of pattern matching processing is used instead of the first inference model 110, thereby reducing the burden associated with the computational processing of the proposed device compared to the proposed device 100.

[0080] Table 111 is an example of first determination data. Table 111 is data in which building information and specification candidates are associated with each other. Storage device 102 is configured to store first database 110DB including table 111. Computing device 101 acquires specification candidates associated with building information from first database 110DB as specification candidates suitable for the building information.

[0081] Third Embodiment Next, a third embodiment will be described with reference to Fig. 15 and Fig. 16. Fig. 15 is a block diagram showing the configuration of a proposed device 100B according to the third embodiment. Fig. 16 is a block diagram for explaining the processing of the proposed device 100B (when a first database 110DB and a second database DBA are used).

[0082] 15, the proposed device 100B according to the third embodiment differs from the proposed device 100A according to the second embodiment in that a second database 120DB is used instead of the second inference model 120, and in that a selection program 131 is used to select necessary information from the first database 110DB and the second database 120DB. In other respects, the configuration of the proposed device 100B is the same as that of the proposed device 100A.

[0083] The second database 120DB includes a second database DBA, a second database DBB, a second database DBC, .... The second database DBA corresponds to index type A, the second database DBB corresponds to index type B, and the second database DBC corresponds to index type C.

[0084] As shown in FIG. 16, the second database DBA has registered therein a table 121 that associates "building information," "specification candidates," and "index values ​​of index type A." One "specification candidate" is registered in table 121 in correspondence with one piece of "building information." The combination of "building information," "specification candidate," and "index value" in one row of table 121 means that the "index value" will be obtained when an air conditioner based on the "specification candidate" is installed in a building based on the "building information."

[0085] Similarly, the second database DBB has registered therein a table associating "building information," "specification candidates," and "index values ​​of index type B." The second database DBC has registered therein a table associating "building information," "specification candidates," and "index values ​​of index type C." Figure 16 shows table 121 registered in the second database DBA as a representative example. Here, the "specification candidates" registered in table 121 are "specification candidates suitable for the building information." As already explained, "specification candidates suitable for the building information" are candidate specifications for air conditioners that are considered desirable to install in a building identified by the building information.

[0086] When the calculation device 101 receives input of building information, it refers to table 111 of first database 110DB and acquires specification candidates 1 to 3 from first database 11DB, as in the second embodiment. The calculation device 101 inputs the acquired specification candidates 1 to 3 into second database DBA. Then, the calculation device 101 refers to table 121 of second database DBA. The calculation device 101 searches table 121 for "index values ​​of index type A" that correspond to the "accepted building information" and the "acquired specification candidates 1 to 3."

[0087] The calculation device 101 retrieves the "index value of index type A" that is found as a result of the search from the second database DBA. Note that while Fig. 16 shows a case where the second database DBA is used, the calculation device 101 executes processing in a similar procedure using the second databases DBA, DBB, DBC, etc., depending on the index type specified by the user.

[0088] According to embodiment 3, compared to embodiment 2, a second database 120DB capable of pattern matching processing is used instead of the second inference model 120, thereby reducing the burden on the proposed device related to computational processing compared to proposed device 100B.

[0089] Table 121 is an example of second judgment data. Table 121 is data in which index values ​​corresponding to index types are associated with building information and specification candidates. The storage device 102 is configured to store a second database 120DB including table 121. The calculation device 101 acquires index values ​​from the second database 120DB as evaluations of specification candidates based on index types.

[0090] Fourth Embodiment Fig. 17 is a block diagram for explaining the configuration of a proposal system 1000 according to a fourth embodiment. The proposal system 1000 includes a proposal device 100C and a communication device 60. The proposal device 100C and the communication device 60 are communicably connected via the Internet 90. The proposal device 100C includes a calculation device 101 and a storage device 102, similar to the proposal devices 100, 100A, and 100B. The proposal device 100C further includes a communication interface 108 that enables communication between the communication device 60 and the calculation device 101.

[0091] The configurations of the arithmetic device 101 and the storage device 102 may be common to the configurations of the arithmetic device 101 and the storage device 102 included in the proposed device 100, may be common to the configurations of the arithmetic device 101 and the storage device 102 included in the proposed device 100A, or may be common to the configurations of the arithmetic device 101 and the storage device 102 included in the proposed device 100B. That is, the proposed device 100C according to the fourth embodiment may have the configuration of any of the proposed devices according to the first to third embodiments. The proposed device 100C may be configured by a server device that builds a cloud system.

[0092] The communication device 60 includes a display device 61, an input interface 62, and a communication interface 63. The communication device 60 may be configured, for example, as a smartphone carried by a user. The communication device 60 may also be configured as a desktop personal computer (PC), a laptop PC, a smart watch, a wearable device, a tablet PC, or the like.

[0093] The display device 61 displays various screens shown in FIGS. 10 and 11 . The user inputs building information and index types to the communication device 60 using the input interface 62. The communication device 60 transmits the building information and index types to the proposal device 100C via the Internet 90. The arithmetic device 101 of the proposal device 100C receives the building information and index types via the communication interface 108. The arithmetic device 101 executes a process of acquiring specification candidates and index values ​​corresponding to the specification candidates from an estimation model or database stored in the storage device 102 using the building information and index types. The details of this process have been described in the first to third embodiments, and therefore will not be repeated here.

[0094] The arithmetic device 101 transmits display information for allowing the user to identify the acquired specification candidates and the evaluations of the specification candidates to the communication device 60 via the Internet 90. The communication device 60 displays a screen such as that illustrated in (4) of Fig. 11 on the display device 61 based on the display information.

[0095] Note that the communication device 60 may accept the input of the building information and the input of the index type at different times, similar to the flowchart shown in Fig. 9. Alternatively, the communication device 60 may accept the input of the building information and the input of the index type at the same time, similar to the flowchart shown in Fig. 12.

[0096] According to the fourth embodiment, the device that calculates the specification candidates and index values ​​(the proposed device 100C) is separated from the device operated by the user (the communication device 60), thereby reducing the processing burden on the device operated by the user (here, the communication device 60). Note that, as shown in FIG. 17 , the proposed system 1000 may be configured so that multiple communication devices 60 can communicate with the proposed device 100C. This allows the computational resources of the proposed device 100C to be used effectively.

[0097] In each embodiment, information consisting of a combination of location, age, purpose, and layout has been introduced as an example of building information. However, the building information may be one or more of these four pieces of information. Alternatively, the building information may be one or more of the information listed in the description of the first embodiment.

[0098] In each embodiment, a combination of an air conditioning method and a capacity range has been introduced as an example of an air conditioner specification. However, the air conditioner specification may be one of the air conditioning method and the capacity range. Alternatively, the air conditioner specification may be one or more of the information listed in the description of the first embodiment.

[0099] 6 illustrates multiple index types. In each embodiment, all of these multiple index band types may be provided to the user as options, or two or more of these multiple index types may be provided to the user as options. Alternatively, the index types provided to the user as options may include at least one of the other examples listed in the description of the first embodiment.

[0100] In each embodiment, an example has been described in which one type of index type is designated by the user. However, the proposed device according to each embodiment may accept designation of multiple index types from the user.

[0101] (Summary) The present embodiment will be summarized below.

[0102] (1) The present disclosure provides a proposal device (100) that proposes specifications for an air conditioner (50) to a user, the proposal device (100) including a computing device (101), a storage device (102), an input interface (103) that accepts input of information from a user, and an output interface (104) that outputs information, the input interface accepting building information regarding a building in which the air conditioner is installed and an index type for evaluating the performance of the air conditioner (steps S101, S105, S201), the storage device accepting first determination data (110) used to acquire candidate specifications for the air conditioner and an index type for acquiring an evaluation of the candidate specifications. The calculation device uses the first determination data to acquire specification candidates suitable for the building information (step S103), and uses the second determination data to acquire an evaluation of the specification candidates based on the index type (step S108), the evaluation of the specification candidates being based on the performance that the air conditioner of the specification candidate is estimated to exhibit when it is installed in the building specified by the building information, and the output interface outputs display information to the display device (20) to allow a user to identify the specification candidates and the evaluation of the specification candidates.

[0103] (2) In the proposed device of paragraph 1, the first judgment data is a first inference model (110) for inferring specification candidates suitable for building information, and the first inference model is trained to infer specification candidates suitable for building information based on learning data including building information and specification candidates, and the calculation device obtains specification candidates suitable for the building information from the first inference model by inputting the building information into the first inference model.

[0104] (3) In the proposed device of paragraph 1 or paragraph 2, the second judgment data is a second inference model (120) for inferring an evaluation of a specification candidate based on an index type, and the second inference model is trained to infer an index value corresponding to the index type based on learning data including building information, specification candidates, and index values ​​corresponding to the index type, and the calculation device inputs the building information and specification candidates into the second inference model to obtain an index value from the second inference model as an evaluation of the specification candidate based on the index type.

[0105] (4) In the proposed device of paragraph 3, the index type includes a first index and a second index, the second inference model includes a first index model (MA) corresponding to the first index and a second index model (MB) corresponding to the second index, and when the index type received by the input interface is the first index, the calculation device uses the first index model to obtain an evaluation of the specification candidate based on the index type, and when the index type received by the input interface is the second index, the calculation device uses the second index model to obtain an evaluation of the specification candidate based on the index type.

[0106] (5) In the proposed device of paragraph 1, the first judgment data is data (111) in which building information and specification candidates are associated with each other, the storage device is configured to store a first database (110DB) including the first judgment data, and the calculation device obtains, from the first database, specification candidates associated with the building information as specification candidates suitable for the building information.

[0107] (6) In the proposed device of paragraph 1 or paragraph 5, the second judgment data is data (121) in which an index value corresponding to an index type is associated with building information and specification candidates, the storage device is configured to store a second database (120DB) containing the second judgment data, and the calculation device obtains the index value from the second database as an evaluation of the specification candidates based on the index type.

[0108] (7) In the proposed device of any one of paragraphs 1 to 6, the building information includes at least one of the layout of the room in which the air conditioner is installed, the location of the building, the age of the building, and the purpose of the building.

[0109] (8) In the proposed device of any one of paragraphs 1 to 7, the specifications include an air conditioning method and a capacity range.

[0110] (9) In the proposed device of any one of paragraphs 1 to 8, the index type includes any one of the following: the COP (Coefficient of Performance) of the air conditioner, the power consumption of the air conditioner, the CO2 emissions of the air conditioner, the installation cost of the air conditioner, and the construction period required to install the air conditioner.

[0111] (10) Another aspect of the present disclosure is a proposal system (1000) that proposes specifications for an air conditioner (50) to a user, the system comprising: a communication device (60); a calculation device (101) configured to communicate with the communication device via a network (90); and a storage device (102) configured to store first determination data used to acquire candidate specifications for the air conditioner and second determination data used to acquire an evaluation of the candidate specifications. When the communication device receives building information about a building in which the air conditioner is to be installed and an index type for evaluating the performance of the air conditioner, the communication device transmits the building information and the index type to the calculation device. The calculation device uses the first determination data to acquire candidate specifications suitable for the building information and uses the second determination data to acquire an evaluation of the candidate specifications based on the index type. The calculation device transmits display information to the communication device to allow a user to identify the candidate specifications and the evaluation of the candidate specifications. The evaluation of the candidate specifications is based on the performance that the candidate air conditioner is estimated to exhibit if it is installed in the building identified by the building information. The communication device displays the display information on the display device (61).

[0112] The embodiments disclosed herein should be considered to be illustrative in all respects and not restrictive. The scope of the present disclosure is defined by the claims, not by the description of the above embodiments, and is intended to include all modifications within the meaning and scope of the claims.

[0113] REFERENCE SIGNS LIST 1 Compressor, 2 Outdoor heat exchanger, 3 Indoor heat exchanger, 4 Expansion valve, 5 Four-way valve, 6, 7 Fan, 8 Refrigerant piping, 10 Input device, 20, 61 Display device, 21 Display, 50 Air conditioner, 51 Outdoor unit, 52 Indoor unit, 60 Communication device, 62, 103 Input interface, 63, 108 Communication interface, 100, 100A, 100B, 100C Proposal device, 101 Arithmetic device, 102 Storage device, 104 Output interface, 110DB First database, 110 First inference model, 111, 121 Table, 120DB Second database, 120 Second inference model, 130 Inference program, 131 Selection program, 201, 301 Computer, 205, 305 Training department, 1000 Proposal system.

Claims

1. A proposal device that proposes air conditioner specifications to a user, comprising: a calculation device; a storage device; an input interface that accepts information input from a user; and an output interface that outputs information, wherein the input interface accepts building information related to a building in which the air conditioner is to be installed and an index type for evaluating the performance of the air conditioner, the storage device is configured to store first judgment data used to obtain candidate specifications for the air conditioner and second judgment data used to obtain an evaluation of the candidate specifications, the calculation device uses the first judgment data to obtain the candidate specifications suitable for the building information, and uses the second judgment data to obtain an evaluation of the candidate specifications based on the index type, the evaluation of the candidate specifications being an evaluation based on performance that the candidate specification air conditioner is estimated to exhibit if it is installed in a building identified by the building information, and the output interface outputs display information to a display device to allow a user to identify the candidate specifications and the evaluation of the candidate specifications.

2. The proposed device described in claim 1, wherein the first judgment data is a first inference model for inferring the specification candidates suitable for the building information, the first inference model is trained to infer the specification candidates suitable for the building information based on learning data including the building information and the specification candidates, and the calculation device obtains the specification candidates suitable for the building information from the first inference model by inputting the building information into the first inference model.

3. The proposal device described in claim 1 or claim 2, wherein the second judgment data is a second inference model for inferring an evaluation of the specification candidate based on the index type, the second inference model is trained to infer an index value corresponding to the index type based on learning data including the building information, the specification candidate, and an index value corresponding to the index type, and the calculation device obtains the index value from the second inference model as an evaluation of the specification candidate based on the index type by inputting the building information and the specification candidate to the second inference model.

4. The proposed device of claim 3, wherein the index type includes a first index and a second index, the second inference model includes a first index model corresponding to the first index and a second index model corresponding to the second index, and the calculation device, when the index type accepted by the input interface is the first index, uses the first index model to obtain an evaluation of the specification candidate based on the index type, and when the index type accepted by the input interface is the second index, uses the second index model to obtain an evaluation of the specification candidate based on the index type.

5. The proposed device described in claim 1, wherein the first judgment data is data in which the building information and the specification candidates are associated, the storage device is configured to store a first database including the first judgment data, and the calculation device obtains from the first database the specification candidates associated with the building information as the specification candidates suitable for the building information.

6. The proposed device described in claim 1 or claim 5, wherein the second judgment data is data in which an index value corresponding to the index type is associated with the building information and the specification candidate, the storage device is configured to store a second database including the second judgment data, and the calculation device obtains the index value from the second database as an evaluation of the specification candidate based on the index type.

7. A proposal device as claimed in any one of claims 1 to 6, wherein the building information includes at least one of the layout of the room in which the air conditioner is to be installed, the location of the building, the age of the building, and the use of the building.

8. The proposed device according to any one of claims 1 to 7, wherein the specifications include an air conditioning method and a capacity band.

9. The proposed device according to any one of claims 1 to 8, wherein the index type includes any one of the following: the COP (Coefficient of Performance) of the air conditioner, the power consumption of the air conditioner, the CO2 emissions of the air conditioner, the installation cost of the air conditioner, and the construction period required for the installation of the air conditioner.

10. A proposal system that proposes air conditioner specifications to a user, comprising: a communications device; a calculation device configured to communicate with the communications device via a network; and a storage device configured to store first judgment data used to obtain candidate specifications for the air conditioner and second judgment data used to obtain an evaluation of the candidate specifications, wherein when the communications device receives building information on a building in which an air conditioner is to be installed and an index type for evaluating the performance of the air conditioner, the communications device transmits the building information and the index type to the calculation device, wherein the calculation device uses the first judgment data to obtain the candidate specifications suitable for the building information, and uses the second judgment data to obtain an evaluation of the candidate specifications based on the index type, and transmits display information to the communications device for allowing a user to identify the candidate specifications and the evaluation of the candidate specifications, wherein the evaluation of the candidate specification is based on performance that is estimated to be exhibited by the air conditioner of the candidate specification if it is installed in a building identified by the building information, and the communications device displays the display information on a display device.

Images