Prediction methods, procedures and prediction systems

Abstract

In the estimation method, first data is acquired, the first data being data related to a state of an apparatus and having a plurality of parameters including a specific parameter (S1). In the estimation method, second data is generated by changing a value of the specific parameter in the acquired first data (S2). In the estimation method, first estimation data is output by inputting the acquired first data to an estimation model that has been learned in such a manner that the above data is input and information indicating a change in the state of the apparatus is output (S3). In the estimation method, second estimation data is output by inputting the second data to the estimation model (S4). In the estimation method, an influence of the specific parameter on the state of the apparatus is estimated based on a relationship between the first estimation data and the second estimation data (S5).

Description

Technical Field

[0001] This disclosure relates to methods for estimating the presumed impact on equipment, etc. Background Technology

[0002] For example, Patent Document 1 discloses an information processing system. This information processing system includes a diagnostic unit and a notification unit. The diagnostic unit performs refrigerator fault diagnosis using a learning model that has been trained to output fault diagnosis results when information representing the operating status of the refrigerator is input. The notification unit outputs information based on the fault diagnosis results from the diagnostic unit from the refrigerator or a terminal device.

[0003] Prior art literature Patent documents Patent Document 1: Japanese Patent Application Publication No. 2021-184132 Summary of the Invention

[0004] The problem that the invention aims to solve This disclosure provides estimation methods, etc., that facilitate the estimation of the impact of parameters related to the state of the equipment on the equipment.

[0005] Methods for solving problems In one aspect of the estimation method disclosed herein, first data is obtained, which is data related to the state of a device and has multiple parameters including a specific parameter. In the estimation method, second data is generated by changing the value of the specific parameter in the obtained first data. In the estimation method, first estimated data is output by inputting the obtained first data into an estimation model that has been learned to take the device-related data as input and output information representing changes in the state of the device. In the estimation method, second estimated data is output by inputting the generated second data into the estimation model. In the estimation method, the influence of the specific parameter on the state of the device is estimated based on the relationship between the first estimated data and the second estimated data.

[0006] One method of this disclosure involves a program that causes one or more processors to execute the presumption method.

[0007] One aspect of this disclosure relates to an estimation system comprising an acquisition unit, a generation unit, a first estimation unit, a second estimation unit, and a third estimation unit. The acquisition unit acquires first data, which is data related to the state of a device and has multiple parameters including specific parameters. The generation unit generates second data by changing the value of the specific parameters in the first data acquired by the acquisition unit. The first estimation unit outputs first estimated data by inputting the first data acquired by the acquisition unit into an estimation model that has been learned to take the data related to the state of the device as input and output information representing changes in the state of the device. The second estimation unit outputs second estimated data by inputting the second data generated by the generation unit into the estimation model. The third estimation unit estimates the influence of the specific parameters on the state of the device based on the relationship between the first estimated data output by the first estimation unit and the second estimated data generated by the second estimation unit.

[0008] Invention Effects The estimation method and the like disclosed herein have the advantage of making it easy to estimate the impact of parameters related to the state of the equipment on the equipment. Attached Figure Description

[0009] Figure 1 This is a diagram illustrating a diagnostic method for assessing the condition of diagnostic equipment.

[0010] Figure 2 This is a diagram illustrating other topics related to diagnostic methods for assessing the condition of diagnostic equipment.

[0011] Figure 3 This is a block diagram illustrating the overall configuration of the estimation system involved in the implementation.

[0012] Figure 4 This is a flowchart illustrating an example of the operation of the estimation system involved in the implementation.

[0013] Figure 5 This is a diagram schematically illustrating the operation of the presumed system involved in the implementation.

[0014] Figure 6 This is a diagram that schematically illustrates a specific example of the operation of the presumed system involved in the implementation. Detailed Implementation

[0015] [1. Basis of this disclosure] First, the inventor's perspective will be explained below.

[0016] In recent years, diagnostic methods have been studied to determine whether a device, such as a refrigerator, has malfunctioned when an abnormality occurs. In particular, these methods aim not only to diagnose whether a malfunction has occurred, but also to determine the cause of the malfunction if it has, or the cause of the abnormality if it has not.

[0017] However, in appliances such as refrigerators, multiple parameters related to the appliance's state can affect its state (e.g., the internal temperature of the refrigerator). These multiple parameters may include those related to user actions on the appliance, such as the number of times the refrigerator door is opened and closed. Furthermore, these multiple parameters may include physical quantities reflecting the appliance's state, such as the temperature of the refrigerator compartment, freezer compartment, or vegetable compartment. Additionally, these multiple parameters may include those related to the appliance's control, such as the frequency of the refrigerator's compressor. Finally, these multiple parameters may include parameters reflecting the appliance's surrounding environment, such as the room temperature of the location where the refrigerator is installed.

[0018] Figure 1 This is a diagram illustrating a diagnostic method for assessing the condition of diagnostic equipment. Figure 1 The example shown illustrates an abnormal situation where the internal temperature rises after a user opens or closes the refrigerator door. Here, as mentioned above, the state of the equipment can be affected by multiple parameters, making it difficult to determine whether opening or closing the refrigerator door is the cause of the temperature rise. For example, the temperature might rise due to a high room temperature in the area where the refrigerator is located. Or, the temperature might rise due to reduced cooling performance caused by overfilling the refrigerator with food. Or, the temperature might rise if the refrigerator door is opened while defrosting or other cooling controls are active. Or, if the refrigerator door is opened and then immediately closed, the internal temperature might not rise at all.

[0019] Here, as a method to estimate which of multiple parameters affects the device, one could consider extracting and verifying samples from a dataset containing information about changes in features and states composed of multiple parameters, where only the specific parameter whose influence is to be estimated differs while all other parameters are identical. However, this method has the following drawbacks.

[0020] Figure 2 This is a diagram illustrating other topics related to diagnostic methods for assessing the condition of diagnostic equipment. Figure 2The data on the left and right sides of the diagram represent an example of a sample where only the number of times the refrigerator door is opened and closed differs as a specific parameter, while all other parameters are identical. The challenge here is that such samples are often difficult to find in actual refrigerator usage environments. Furthermore, even if such samples are found, preparing a sufficient number of samples to determine whether the opening and closing of the refrigerator door is the cause of its impact is extremely difficult.

[0021] Furthermore, as a method for diagnosing which of multiple parameters is causing the impact on the device, a presumptive model could be constructed. This model, for example, learns by taking data containing multiple parameters as input and outputting information indicating the impact of the specific parameter on the device's state. However, this method suffers from the problem that learning is difficult because accurate data representing the degree of influence of a specific parameter in changes in the device's state is unavailable.

[0022] In view of the above, the inventors have completed this disclosure.

[0023] Hereinafter, embodiments will be described in detail with reference to the accompanying drawings. Furthermore, the embodiments described below are general or specific examples. The numerical values, shapes, materials, constituent elements, the arrangement and connection methods of constituent elements, steps, and the order of steps shown in the following embodiments are merely examples and are not intended to limit this disclosure. In addition, constituent elements not described in the independent claims in the following embodiments will be described as arbitrary constituent elements.

[0024] Furthermore, the figures are schematic diagrams and not necessarily strictly representational. Also, in each figure, substantially identical components are labeled the same, and sometimes repeated descriptions are omitted or simplified.

[0025] [2. Composition] The following is for reference Figure 3 The description includes the overall configuration of the estimation system 1 involved in the implementation method. Figure 3 This is a block diagram showing the overall configuration of the estimation system 1 included in the embodiments. Figure 3 In addition to the estimation system 1, device 2 is also illustrated. Estimation system 1 is a system used to diagnose the condition of device 2.

[0026] Device 2 is, for example, a device installed in the residential facility of a user utilizing the presumption system 1. Device 2 is, for example, a home appliance. More specifically, device 2 is, for example, a device capable of connecting to a network N1 such as the Internet, i.e., a so-called IoT (Internet of Things) device. Specifically, device 2 is, for example, a refrigerator, an air conditioner, a washing machine, a robot vacuum cleaner, a rice cooker, or a microwave oven. In this embodiment, device 2 is a refrigerator, a continuously operating device, in other words, a device that is essentially always running. Furthermore, device 2 can also be a device that operates intermittently. Additionally, if device 2 is equipped with the presumption system 1 as described below, it may not be an IoT device.

[0027] System 1 is presumably a computer that includes a processor (e.g., a microprocessor) and memory. The memory, such as ROM (Read-Only Memory) and RAM (Random Access Memory), is capable of storing programs executed by the processor.

[0028] For example, the presumed system 1 can be a single-unit computer (device) or a system composed of multiple computers. For example, the presumed system 1 can also be a server. Furthermore, the components of the presumed system 1 can be configured on a single server or distributed across multiple servers. Additionally, for example, the presumed system 1 can also be mounted on device 2.

[0029] The estimation system 1 includes an acquisition unit 11, a generation unit 12, a first estimation unit 13, a second estimation unit 14, and a third estimation unit 15. The acquisition unit 11, the generation unit 12, the first estimation unit 13, the second estimation unit 14, and the third estimation unit 15 are implemented by processors or the like that that execute programs stored in memory.

[0030] The acquisition unit 11 acquires first data. Here, the first data is data related to the state of the device 2 and has multiple parameters including specific parameters. In this embodiment, the first data is data related to the state of the refrigerator and includes parameters such as the room temperature of the location where the refrigerator is installed, the internal temperature of the refrigerator (temperature of the refrigerator compartment, temperature of the vegetable compartment, and temperature of the freezer compartment, etc.), the frequency of the compressor, the number of times the refrigerator door is opened and closed, the opening and closing status of the damper, and the defrosting operation status. These parameters are measured in the device 2 at a certain period (e.g., 5 minutes, etc.). Furthermore, the number of times the refrigerator door is opened and closed is the number of times it is opened and closed within a certain period.

[0031] In this implementation, there is no collinearity between the specific parameter and one or more parameters other than the specific parameter among the plurality of parameters included in the first data. To confirm this, a method may be considered, for example, to confirm that the absolute value of the correlation coefficient between the specific parameter and one or more parameters is below a predetermined value.

[0032] Furthermore, in this embodiment, the specific parameter is a value related to the user's actions on the device 2 using the estimation system 1. More specifically, in this embodiment, the specific parameter is a characteristic quantity based on at least one of the number of times the refrigerator door is opened and closed and the opening and closing time of the door. In this embodiment, only the number of times the refrigerator door is opened and closed is used as this characteristic quantity. Furthermore, in this embodiment, one or more parameters include at least one of a physical quantity representing the state of the device 2, a value related to the control of the device 2, and a physical quantity representing the surrounding state of the device 2. More specifically, in this embodiment, one or more parameters include the internal temperature of the refrigerator (temperature of the refrigerator compartment, temperature of the vegetable compartment, and temperature of the freezer compartment, etc.) as a physical quantity representing the state of the device 2, and the compressor frequency as a value related to the control of the device 2. Furthermore, in this embodiment, one or more parameters include the room temperature of the location where the refrigerator is installed as a physical quantity representing the surrounding state of the device 2.

[0033] In this implementation, the acquisition unit 11 acquires first data from the raw data. In other words, the acquisition unit 11 extracts feature quantities from the raw data to serve as input data for the estimation model 16 described later. Here, the raw data is data measured periodically in the device 2 (hereinafter referred to as "measurement data"), which is a collection of data accumulated before the estimation system 1 performs the diagnosis. The estimation system 1 can receive the raw data from a cloud containing the measurement data of the device 2 via wireless or wired communication through a communication interface via network N1.

[0034] The acquisition unit 11 may, for example, acquire measurement data at a time specified by the user using the estimation system 1 or the most recent measurement data at that time from the raw data as first data. Alternatively, the acquisition unit 11 may also acquire the latest measurement data at the time when the estimation system 1 performs the diagnosis from the raw data as first data.

[0035] The generation unit 12 generates second data by changing the value of a specific parameter in the first data acquired by the acquisition unit 11. The second data is data in which only the value of the specific parameter changes relative to the value of the specific parameter in the first data, while one or more parameters other than the specific parameter are all identical to one or more parameters in the first data. In other words, the second data is virtual data generated from the first data.

[0036] In this embodiment, the generation unit 12 generates the second data by changing a specific parameter in the first data obtained by the acquisition unit 11, namely the number of times the refrigerator door is opened and closed, to "0". Furthermore, the value of the specific parameter in the second data does not have to be "0", as long as it differs from the value of the specific parameter in the first data.

[0037] The first estimation unit 13 inputs the first data acquired by the acquisition unit 11 into the estimation model 16 and outputs the first estimation data. Here, the estimation model 16 is a machine learning model that has been learned in a way that takes the data with multiple parameters (in other words, measurement data) as input and outputs information (predicted value) representing the change in the state of the device 2. The estimation model 16 is constructed, for example, using LightGBM (registered trademark) or a neural network.

[0038] The estimation model 16 is constructed through supervised learning using a learning dataset consisting of a large number of samples. Here, the learning dataset includes measurement data as input data and changes in the state of device 2 as correct data. In this embodiment, the measurement data as input data is data measured periodically within the refrigerator. Furthermore, in this embodiment, the changes in the state of device 2, which are considered correct data, are changes in the internal temperature of the refrigerator. More specifically, the change in the internal temperature of the refrigerator is the difference between the internal temperature (e.g., the temperature of the refrigerator compartment) in the measurement data and the internal temperature in the measurement data of the next period.

[0039] Therefore, in this embodiment, the estimation model 16 is trained by taking the data measured by the refrigerator as input data and outputting information representing the change in the refrigerator's internal temperature from the time the data was measured to a time after a predetermined period (equivalent to a certain period). Therefore, in this embodiment, the first estimation data output by the first estimation unit 13 represents the change in the refrigerator's internal temperature from the time the first data was measured to a time after the predetermined period.

[0040] The second estimation unit 14 inputs the second data generated by the generation unit 12 into the estimation model 16 and outputs second estimated data. Here, the estimation model 16 used by the first estimation unit 13 and the estimation model 16 used by the second estimation unit 14 are the same model. Therefore, in this embodiment, the second estimated data output by the second estimation unit 14 represents the change in the refrigerator's internal temperature from the time the second data was measured to a time after a predetermined period. Furthermore, the second data is virtual data that was not actually measured, so the second estimated data is data assuming that the second data was actually measured. Here, the time when the second data was measured can be replaced with the time when the first data was measured.

[0041] The third estimation unit 15 estimates the influence of a specific parameter on the state of the device 2 based on the relationship between the first estimation data obtained by comparing the first estimation data output by the first estimation unit 13 and the second estimation data output by the second estimation unit 14.

[0042] In this embodiment, the third estimation unit 15 estimates the aforementioned effect caused by the difference between the values ​​of a specific parameter in the first data and the second data, based on the relationship between the first and second estimated data. More specifically, in this embodiment, the third estimation unit 15 calculates the difference between the first and second estimated data, that is, the difference in the temperature change inside the refrigerator based on data where the specific parameter (here, the number of times the refrigerator door is opened and closed) is different. Furthermore, the third estimation unit 15 considers this calculated difference as the effect of the difference between the values ​​of the specific parameter in the first data and the specific parameter in the second data, i.e., the difference in the number of times the refrigerator door is opened and closed between the original data and the virtual data, on the temperature change inside the refrigerator. Thus, the third estimation unit 15 estimates that the difference in the number of times the refrigerator door is opened and closed has an effect equivalent to the difference between the first and second estimated data on the temperature change inside the refrigerator.

[0043] [3. Action] The following is for reference Figure 4 and Figure 5 The operation (i.e., estimation method) of the estimation system 1 involved in the implementation method is explained. Figure 4 This is a flowchart illustrating an example of the operation of the estimation system 1 according to the embodiment. Figure 5 This is a diagram schematically illustrating the operation of the estimation system 1 involved in the embodiment. Figure 4 (a) represents a series of processes presuming the actions of system 1. Figure 4 (b) represents a specific example of the process for presuming the status of the diagnostic equipment in System 1.

[0044] First, such as Figure 4 (a) and Figure 5 As shown, the acquisition unit 11 of the estimation system 1 acquires the first data (S1). In the embodiment, as described above, the acquisition unit 11 acquires the first data from the original data (in other words, the acquisition unit 11 extracts feature quantities from the original data to be used as input data for the estimation model 16).

[0045] Next, as Figure 4 (a) and Figure 5 As shown, the generation unit 12 of the estimation system 1 generates second data by changing the value of a specific parameter in the first data acquired by the acquisition unit 11 (S2). That is, the generation unit 12 only changes the value of the specific parameter in the first data that is desired to have an estimated influence to a virtual specific parameter that is desired to be compared, without changing the values ​​of the remaining one or more parameters, thereby generating the second data. In this embodiment, as described above, the second data is generated by changing the value of the specific parameter in the first data to "0".

[0046] Next, as Figure 4 (a) and Figure 5 As shown, the first estimation unit 13 of the estimation system 1 inputs the first data acquired by the acquisition unit 11 into the estimation model 16 and outputs the first estimation data (S3). Furthermore, as... Figure 4 (a) and Figure 5 As shown, the second estimation unit 14 of the estimation system 1 inputs the second data generated by the generation unit 12 into the estimation model 16 and outputs the second estimation data (S4). In addition, steps S3 and S4 can be executed out of order, in reverse order, or in parallel.

[0047] Next, as Figure 4 (a) and Figure 5 As shown, the third estimation unit 15 of the estimation system 1 estimates the influence of a specific parameter on the state of the device 2 based on the relationship between the first estimation data output by the first estimation unit 13 and the second estimation data output by the second estimation unit 14 (S5). Specifically, as... Figure 4 As shown in (b), in step S5, the third estimation unit 15 first calculates the difference between the value of the specific parameter of the first estimation data and the value of the specific parameter of the second estimation data (S51). Then, based on the calculated difference, the third estimation unit 15 estimates the effect of the specific parameter on the state of the device 2 (S52).

[0048] Figure 6 This is a diagram that schematically illustrates a specific example of the operation of the estimation system 1 involved in the embodiment. Figure 6 The example shown is when device 2 is a refrigerator. Furthermore, in Figure 6 In the example shown, the specific parameter in the first data is the number of times the refrigerator door is opened and closed, and one or more parameters in the first data include the room temperature of the location where the refrigerator is set, the temperature of the refrigerator compartment, the temperature of the freezer compartment, the temperature of the vegetable compartment, and the frequency of the compressor.

[0049] exist Figure 6 In the example shown, the generation unit 12 of the estimation system 1 generates the second data by changing the number of times the refrigerator door was opened and closed in the first data from "10 times" to "0 times". Furthermore, the first estimation unit 13 of the estimation system 1 inputs the first data into the estimation model 16 and outputs first estimated data indicating that the internal temperature of the refrigerator (here, the temperature of the refrigerator compartment) will rise by 3 degrees Celsius after 5 minutes. Furthermore, the second estimation unit 14 of the estimation system 1 inputs the second data into the estimation model 16 and outputs second estimated data indicating that the internal temperature of the refrigerator will rise by 1 degree Celsius after 5 minutes. Then, the third estimation unit 15 of the estimation system 1 estimates, based on the difference between the first and second estimated data, that the opening and closing of the refrigerator door causes the internal temperature of the refrigerator to rise by 2 degrees Celsius.

[0050] [4. Effects, etc.] As described above, in the estimation system 1 (estimation method) according to the embodiment, first data and second data, whose values ​​differ only from each other for specific parameters, are input into the same estimation model 16, thus enabling the estimation of the effect of only specific parameters on changes in the state of the device 2. Therefore, the estimation system 1 according to the embodiment has the advantage of easily estimating which of a plurality of parameters affects the device 2.

[0051] Furthermore, in the estimation system 1 according to the embodiment, the second data is generated by changing a specific parameter in the acquired first data. That is, the estimation system 1 according to the embodiment generates virtual measurement data in which only a specific parameter is changed relative to the measurement data, and therefore has the advantage that there is no problem even if there is no sample as described in [1. The basis of this disclosure] where only the specific parameter is different and all other parameters are the same.

[0052] Furthermore, in the estimation system 1 according to the embodiment, it is sufficient to prepare an estimation model 16 that has been learned in a manner that takes data with multiple parameters (in other words, measurement data) as input and outputs information representing changes in the state of device 2. Therefore, the estimation system 1 according to the embodiment has the advantage of being easier to prepare a large training dataset required for learning the estimation model 16 and easier to construct the estimation model 16 compared to the case of constructing the estimation model described in [1. Basis of this disclosure]. This is because in the estimation system 1 according to the embodiment, when learning the estimation model 16, it is sufficient to prepare a training dataset that includes measurement data as input data and changes in the state of device 2 as correct data, and it is not necessary to prepare a training dataset that includes correct data representing the degree of influence of specific parameters in changes in the state of device 2.

[0053] [5. Other Implementation Methods] As described above, embodiments have been illustrated as examples of the technology disclosed in this application. However, the technology in this disclosure is not limited thereto, and can also be applied to embodiments with appropriate modifications, substitutions, additions, or omissions. Furthermore, new embodiments can be formed by combining the constituent elements described in the above embodiments.

[0054] In the above embodiment, when device 2 is a refrigerator, the specific parameter is the number of times the refrigerator door is opened and closed, but it is not limited to this. For example, the specific parameter may also be the contents of the refrigerator (the total weight of food, etc., stored in the refrigerator), the frequency of the compressor, the opening and closing of the damper, or the presence or absence of defrosting, etc. Furthermore, the specific parameter is not limited to one type, but may be multiple types. In addition, the specific parameter may be appropriately changed by the user using the estimation system 1.

[0055] For example, when the specific parameter is the amount of contents inside the refrigerator, the third estimation unit 15 can estimate the effect of the amount of contents inside the refrigerator on the change of the internal temperature. Furthermore, for example, when the specific parameter is at least one of the compressor frequency, the opening and closing of the damper, and the presence or absence of defrosting, the third estimation unit 15 can estimate the effect of these specific parameters on the change of the internal temperature.

[0056] Furthermore, the processing order described in the above embodiments is only one example. The order of multiple processes can be changed, and multiple processes can also be executed in parallel. In addition, the processing performed by a specific processing unit can also be performed by other processing units. Furthermore, some of the digital signal processing described in the above embodiments can also be implemented by analog signal processing.

[0057] For example, this disclosure can also be implemented as a program for causing a computer (processor) to perform the steps included in the presumed method. Furthermore, this disclosure can also be implemented as a non-transitory computer-readable recording medium such as a CD-ROM containing the program.

[0058] For example, when this disclosure is implemented by a program (software), each step is executed by utilizing the computer's hardware resources such as the CPU, memory, and input / output circuits. That is, each step is executed by the CPU retrieving data from memory or input / output circuits for computation, or outputting the computation results to memory or input / output circuits.

[0059] Furthermore, in the above embodiments, it is assumed that each component of the system 1 can be constructed by dedicated hardware, or it can be implemented by executing software programs suitable for each component. Each component can also be implemented by a program execution unit such as a CPU or processor reading and executing software programs recorded on recording media such as hard disks or semiconductor memories.

[0060] The functions of the presumed system 1 described in the above embodiments are typically implemented by an LSI as an integrated circuit. These can be implemented on individual chips or partially or entirely within a single chip. Furthermore, the integrated circuit implementation is not limited to LSIs; it can also be implemented using dedicated circuits or general-purpose processors. Alternatively, an FPGA (Field-Programmable Gate Array) programmable after LSI manufacturing, or a reconfigurable processor capable of reconfiguring the connections or settings of the internal circuitry units of the LSI, can be utilized.

[0061] Furthermore, if an integrated circuit technology to replace LSI emerges due to advancements in semiconductor technology or other derived technologies, then this technology can certainly be used to realize the integrated circuitization of each component included in the presumed system 1.

[0062] Furthermore, this disclosure also includes various modifications that can be conceived by those skilled in the art to the embodiments, and ways of implementing them by arbitrarily combining the constituent elements and functions of the embodiments without departing from the spirit of this disclosure.

[0063] (Summarize) As described above, in the estimation method involved in the first approach, first data is obtained, which is data related to the state of device 2 and has multiple parameters including specific parameters (S1). In this estimation method, second data is generated by changing the value of the specific parameters in the obtained first data (S2). In this estimation method, first estimated data is output by inputting the obtained first data into an estimation model 16 that has been learned in a manner that takes the above data as input and outputs information representing changes in the state of device 2 (S3). In this estimation method, second estimated data is output by inputting the second data into the estimation model 16 (S4). In this estimation method, the influence of the specific parameters on the state of device 2 is estimated based on the relationship between the first estimated data and the second estimated data (S5).

[0064] Therefore, since the first and second data, whose values ​​differ only in specific parameters, are input into the same estimation model 16, it has the advantage of being able to easily estimate the influence of parameters (specific parameters) related to the state of device 2 on changes in the state of device 2.

[0065] Furthermore, as the estimation method involved in the second method, in the first method, in the processing of the estimation effect (S5), as the relationship between the first estimation data and the second estimation data, based on the difference between the first estimation data and the second estimation data, the effect caused by the difference between the value of a specific parameter in the first data and the value of a specific parameter in the second data is estimated (S51, S52).

[0066] Therefore, it has the advantage of being able to easily and accurately estimate the impact of the difference in a specific parameter on the state of device 2.

[0067] Furthermore, as an estimation method involved in the third approach, in the first or second approach, the specific parameter is a value related to the user's actions on the device 2. The more than one parameter includes at least one of a physical quantity representing the state of the device 2, a value related to the control of the device 2, and a physical quantity representing the surrounding state of the device 2.

[0068] Therefore, it has the advantage of being able to easily estimate the impact of the user's actions on the device 2 on the state of the device 2.

[0069] Furthermore, as a presumption method involved in the fourth method, in any of the first to third methods, device 2 is a refrigerator. The change in the state of device 2 is the change in the internal temperature.

[0070] Therefore, it has the advantage of making it easy to estimate the effect of specific parameters on the temperature changes inside the refrigerator.

[0071] Furthermore, as an estimation method involved in the fifth method, in the fourth method, the specific parameter is a characteristic quantity based on at least one of the number of times the refrigerator door is opened and closed and the opening and closing time of the door.

[0072] Therefore, it has the advantage of being able to easily estimate the effect of at least one of the number of times the refrigerator door is opened and closed and the opening and closing time on the temperature change inside the refrigerator.

[0073] Furthermore, the program involved in method 6 causes one or more processors to execute the presumed method of any of methods 1 through 5.

[0074] Therefore, it has the advantage of providing a program that can easily estimate the influence of parameters related to the state of device 2 on the state of device 2.

[0075] Furthermore, the estimation system 1 involved in the seventh method includes an acquisition unit 11, a generation unit 12, a first estimation unit 13, a second estimation unit 14, and a third estimation unit 15. The acquisition unit 11 acquires first data, which is data related to the state of the device 2 and has multiple parameters including specific parameters. The generation unit 12 generates second data by changing the value of the specific parameters in the first data acquired by the acquisition unit 11. The first estimation unit 13 outputs first estimated data by inputting the first data acquired by the acquisition unit 11 into an estimation model 16 that has been learned to take the aforementioned data as input and output information representing changes in the state of the device 2. The second estimation unit 14 generates second estimated data by inputting the second data generated by the generation unit 12 into the estimation model 16. The third estimation unit 15 estimates the influence of the specific parameters on the state of the device 2 based on the relationship between the first estimated data generated by the first estimation unit 13 and the second estimated data generated by the second estimation unit 14.

[0076] Therefore, it has the advantage of providing an estimation system 1 that can easily estimate the influence of parameters related to the state of device 2 on device 2.

[0077] Industrial applicability This disclosure can be applied to systems, for example, for diagnosing the condition of equipment such as refrigerators.

[0078] Marker description 1. Presumption System 11 Acquisition Department 12 Generation Department 13 First Presumption 14. Presumption Part 2 15. Presumption Part 3 16. Presumed Model 2 Equipment N1 network.

Claims

1. A method of estimation, Obtain the first data, which is data related to the device's status and has multiple parameters including specific parameters. The second data is generated by changing the value of the specific parameter in the obtained first data. The first inference data is output by inputting the acquired first data into an inference model that has been learned in a manner that takes the data related to the state of the device as input and outputs information representing changes in the state of the device. By inputting the generated second data into the inference model, the second inference data is output. Based on the relationship between the first estimated data and the second estimated data, the impact of the specific parameter on the state of the device is estimated.

2. The estimation method as described in claim 1, In the process of presuming the effect, the effect of the difference between the value of the specific parameter in the first data and the specific parameter in the second data is presumed as the relationship based on the difference between the first presumed data and the second presumed data.

3. The estimation method as described in claim 1 or 2, The specific parameter is a value related to the user's actions on the device. The one or more parameters include at least one of a physical quantity representing the state of the device, a value related to the control of the device, and a physical quantity representing the surrounding state of the device.

4. The estimation method as described in claim 1 or 2, The device is a refrigerator. The change in the state of the equipment is the change in the temperature inside the chamber.

5. The estimation method as described in claim 4, The specific parameter is a characteristic quantity based on at least one of the number of times the refrigerator door is opened and closed and the opening and closing time of the door.

6. A program that causes one or more processors to execute the presumed method as described in claim 1 or 2.

7. A presumption system, comprising: The acquisition unit acquires first data, which is data related to the state of the device and has multiple parameters including specific parameters; The generation unit generates second data by changing the value of the specific parameter in the first data obtained by the acquisition unit; The first estimation unit outputs first estimation data by inputting the first data obtained by the acquisition unit into an estimation model that has been learned in a manner that takes the data related to the state of the device as input and outputs information representing the change in the state of the device. The second estimation unit inputs the second data generated by the generation unit into the estimation model and outputs second estimation data; and The third estimation unit estimates the impact of the specific parameter on the state of the device based on the relationship between the first estimation data output by the first estimation unit and the second estimation data generated by the second estimation unit.

Images