Determination device, determination method, and determination program
By using neural network models with the same working principle but different models for data correction, the problem of the lack of learning data in the new equipment is solved, and high-precision operation information determination and control of equipment without application performance is achieved.
Patent Information
- Application Number
- CN202280095400.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2022-06-20
- Publication Date
- 2024-12-13
AI Technical Summary
The newly developed, useless performance equipment lacks learning data for building neural network models, resulting in the inability to effectively determine the operation-related information.
A neural network model is used for existing devices with the same working principle but different models. Data correction is performed by obtaining input data and calculation coefficients, and the corrected data is used for judgment.
It realizes high-precision operation information determination of equipment without application performance to ensure the normal operation and control of the equipment.
Smart Images
Figure CN119137611A_ABST
Abstract
Description
Technical Field
[0001] Embodiments of the present invention relate to a determination device, a determination method, and a determination program. Background Art
[0002] In recent years, with the development of IoT (Internet of Things) technology, huge amounts of data can be collected in various fields. In the industrial field, a technology has been developed to use the data collected using IoT technology to build a neural network model (NN model) for determining information related to the operation of equipment.
[0003] By using the NN model in each device, it is possible to accurately and efficiently determine information used for control, etc. by taking into account a large amount of information involved in operation.
[0004] Prior art literature
[0005] Patent Literature
[0006] Patent Document 1: Japanese Patent No. 6791429 Summary of the invention
[0007] In order to construct an NN model as described above, a plurality of data acquired when the corresponding equipment is used is required as learning data. As a technique for coping with the situation where it is difficult to acquire a large amount of data when the equipment is used, a method has been proposed to use the existing NN model to suppress the learning data used to construct a new NN model to a small amount.
[0008] However, there is a problem that such a method cannot be used for new product equipment that has no operational records before being put on the market, because there is no data that can be used as learning data.
[0009] The present invention has been made in view of the above-mentioned situation, and an object of the present invention is to provide a determination device, a determination method and a determination program capable of determining information related to the operation of equipment with no operational records using a neural network model.
[0010] The determination device of the present invention is completed in view of the above situation, and is characterized by comprising: a model information storage unit, which stores a neural network model, and the neural network model is constructed for the operation of a second device that operates according to the same working principle as the first device as the control object and has a different model; an information acquisition unit, which acquires input data of a first item related to the device that operates according to the working principle, a first coefficient used to calculate output data of a second item based on the input data of the first item in the first device, and a second coefficient used to calculate output data of the second item based on the input data of the first item in the second device; a determination unit, which determines the output data of the second item related to the first device based on the input data of the first item using the neural network model stored in the model information storage unit; and a correction unit, which corrects the corresponding data by multiplying the input data of the first item or the output data of the second item used in the determination unit by the ratio of the first coefficient to the second coefficient.
[0011] In addition, in the determination method of the present invention, a neural network model is stored, which is constructed for the operation of a second device that operates according to the same working principle as the first device as the control object and has a different model. The input data of the first item related to the device that operates according to the working principle, the first coefficient used to calculate the output data of the second item based on the input data of the first item in the first device, and the second coefficient used to calculate the output data of the second item based on the input data of the first item in the second device are obtained, the input data of the first item is multiplied by the ratio of the first coefficient to the second coefficient to be corrected, the output data of the second item related to the first device is determined based on the corrected input data of the first item using the neural network model, or the output data of the second item is determined based on the input data of the first item using the neural network model, and then the output data of the determined second item is multiplied by the ratio of the first coefficient to the second coefficient to be corrected in a manner corresponding to the first device.
[0012] In addition, the determination program of the present invention enables a computer to perform the following functions: storing a neural network model, which is constructed for the operation of a second device that operates according to the same working principle as the first device as the control object and has a different model; obtaining input data of a first item related to the device that operates according to the working principle, a first coefficient for calculating output data of a second item based on the input data of the first item in the first device, and a second coefficient for calculating output data of the second item based on the input data of the first item in the second device; correcting the input data of the first item by multiplying the ratio of the first coefficient to the second coefficient, using the neural network model to determine the output data of the second item related to the first device based on the corrected input data of the first item, or using the neural network model to determine the output data of the second item based on the input data of the first item, and then correcting the determined output data of the second item by multiplying the ratio of the first coefficient to the second coefficient in a manner corresponding to the first device. BRIEF DESCRIPTION OF THE DRAWINGS
[0013] Figure 1 It is a block diagram showing the configuration of an air conditioner using a first control device as a determination device according to the first embodiment of the present invention.
[0014] Figure 2 This is a block diagram showing the configuration of an air conditioner corresponding to the neural network model used in the determination device according to the first and second embodiments of the present invention.
[0015] Figure 3 This is a flowchart showing the flow of processing executed by the first control device as the determination device according to the first embodiment of the present invention.
[0016] Figure 4 It is a block diagram showing the configuration of an air conditioner using a third control device as a determination device according to a second embodiment of the present invention.
[0017] Figure 5 This is a flowchart showing the flow of processing executed by the third control device as the determination device according to the second embodiment of the present invention.
[0018] Figure 6 (a) is a graph showing positions as a set of points, wherein the positions show states corresponding to the electronic expansion valve opening prediction values determined by the neural network model constructed for the operation of the air conditioner of the model with actual operation performance without data correction for various operating conditions and the electronic expansion valve opening actual measured values measured in the air conditioner of the model to be verified; Figure 6(b) is a graph showing positions as a set of points, wherein the positions show the electronic expansion valve opening prediction values determined by correcting data for various operating conditions using a neural network model constructed for the operation of air conditioners of models with actual operating performance, and the states corresponding to the electronic expansion valve opening actual measured values in the air conditioners of the verification model. DETAILED DESCRIPTION
[0019] 《First Implementation Method》
[0020] An air conditioner using a determination device according to a first embodiment of the present invention is described. In the first embodiment, in an air conditioner of a first type that is newly developed and has no operational record as a first device, a neural network model (NN model) constructed for the operation of an air conditioner of a second type that has an operational record as a second device is used to operate the device in the air conditioner.
[0021] <Configuration of Air Conditioner Using Determination Device According to First Embodiment>
[0022] Reference Figure 1 The configuration of an air conditioner using a determination device according to a first embodiment of the present invention will be described. The air conditioner 1 according to this embodiment is a first model air conditioner that has been newly developed and has no operational records.
[0023] The air conditioner 1 includes a first compressor 11 constituting a first refrigeration cycle 10 , a first evaporator 12 , a first electronic expansion valve 13 , a first condenser 14 , and a first control device 20 as a determination device.
[0024] The first compressor 11 sucks in refrigerant, compresses the sucked refrigerant and discharges it in a high-temperature and high-pressure state. The high-temperature and high-pressure refrigerant discharged from the first compressor 11 dissipates heat and condenses in the first condenser 14 through heat exchange. The condensed refrigerant is decompressed by the first electronic expansion valve 13 and becomes a low-temperature and low-pressure state. The low-temperature and low-pressure refrigerant absorbs heat and evaporates in the first evaporator 12 through heat exchange. The evaporated and gasified refrigerant returns to the first compressor 11 via a liquid accumulator (not shown).
[0025] The first control device 20 includes a first storage unit 21, a first input unit 22, and a first CPU 23, and controls the devices 11, 12, 13, and 14 in the first refrigeration cycle 10. The first storage unit 21 includes a first setting information storage unit 211 and a first NN model information storage unit 212.
[0026] The first setting information storage unit 211 stores setting information related to each of the devices 11 , 12 , 13 , and 14 in the first refrigeration cycle 10 .
[0027] The first NN model information storage unit 212 stores the NN model M constructed by learning the operation data for the refrigeration cycle used in the second type of air conditioner having actual operation records.
[0028] The NN model M is information constructed by performing machine learning using a neural network based on various data collected from various devices in the air conditioner using IoT technology when the second model air conditioner is used. The neural network used to construct the NN model M includes three layers: an input layer, an intermediate layer, and an output layer. The structure of the second model air conditioner is described below.
[0029] The first input unit 22 inputs instructions related to the operation of the devices 11, 12, 13, and 14 in the first refrigeration cycle 10 and information used in the processing corresponding to the instructions. Specifically, the first input unit 22 inputs input data of the first item related to the operation of the devices in the first refrigeration cycle 10 and an instruction to output output data of the second item related to the operation of the devices in the first refrigeration cycle 10 from the input data of the first item. And the first input unit 22 inputs a second coefficient as information used in the processing corresponding to the instruction, and the second coefficient is used to calculate the output data of the second item based on the input data of the first item related to the refrigeration cycle used in the second model air conditioner.
[0030] The first CPU 23 includes a first information acquisition unit 231 , a first correction unit 232 , a first determination unit 233 , and a first operation control unit 234 .
[0031] The first information acquisition unit 231 acquires the command input from the first input unit 22 and the information used in the process corresponding to the command. The first information acquisition unit 231 acquires the information used in the process corresponding to the acquired command from the information stored in the first setting information storage unit 211.
[0032] The first correction unit 232 corrects the input data of the first item related to the operation of the first refrigeration cycle 10 acquired by the first information acquisition unit 231 by multiplying the input data of the first item by the ratio of the first coefficient to the second coefficient.
[0033] The first determination unit 233 determines the output data of the second item related to the operation of the first refrigeration cycle 10 using the NN model M based on the input data of the first item corrected by the first correction unit 232 .
[0034] The first operation control unit 234 controls the operation of each device 11 , 12 , 13 or 14 in the first refrigeration cycle 10 based on the output data of the second item related to the first refrigeration cycle 10 determined by the first determination unit 233 and the information stored in the first setting information storage unit 211 .
[0035] Next, refer to Figure 2 The configuration of the second type of air conditioner will be described. The second type of air conditioner 2 operates on the same operating principle as the first type of air conditioner 1, but uses different components and the like.
[0036] The air conditioner 2 includes a second compressor 31, a second evaporator 32, a second electronic expansion valve 33, a second condenser 34, and a second control device 40 that constitute a second refrigeration cycle 30. The functions of the devices 31, 32, 33, and 34 that constitute the second refrigeration cycle 30 are the same as the functions of the devices 11, 12, 13, and 14 that constitute the first refrigeration cycle 10, and therefore detailed descriptions thereof are omitted.
[0037] The second control device 40 includes a second storage unit 41, a second input unit 42, and a second CPU 43, and controls the devices 31, 32, 33, and 34 in the second refrigeration cycle 30. The second storage unit 41 includes a second setting information storage unit 411 and a second NN model information storage unit 412.
[0038] The second setting information storage unit 411 stores setting information related to each device 31, 32, 33, and 34 in the second refrigeration cycle 30. The second NN model information storage unit 412 stores the NN model M described above.
[0039] The second input unit 42 inputs instructions related to the operation of the devices 31, 32, 33, and 34 in the second refrigeration cycle 30 and information used in the processing corresponding to the instructions. Specifically, the second input unit 42 inputs input data of a first item related to the operation of the devices in the second refrigeration cycle 30, and an instruction to output output data of a second item related to the operation of the devices in the second refrigeration cycle 30 from the input data of the first item.
[0040] The second CPU 43 includes a second information acquisition unit 431 , a second determination unit 432 , and a second operation control unit 433 .
[0041] The second information acquisition unit 431 acquires the command input from the second input unit 42 and the information used in the process corresponding to the command. The second information acquisition unit 431 acquires the information used in the process corresponding to the acquired command from the information stored in the second setting information storage unit 411.
[0042] The second determination unit 432 determines the output data of the second item related to the operation of the second refrigeration cycle 30 using the NN model M based on the input data of the first item acquired by the second information acquisition unit 431 .
[0043] The second operation control unit 433 controls the operation of each device 31 , 32 , 33 or 34 in the second refrigeration cycle 30 based on the output data of the second item related to the second refrigeration cycle 30 determined by the second determination unit 432 and the information stored in the second setting information storage unit 411 .
[0044] <Operation of Air Conditioner Using Determination Device According to First Embodiment>
[0045] The operation of the air conditioner 1 of the present embodiment will be described. Here, the operation of the air conditioner 1 will be described by taking a regression model as an example, which uses the operating frequency of the motor used in the first compressor 11 in the first refrigeration cycle 10 (compressor frequency Hz) and the specific volume of the suction gas of the first compressor 11 (compressor suction gas specific volume v) as the explanatory variables, and uses the refrigerant circulation amount G in the first refrigeration cycle 10 as the output data of the second item as the target variable.
[0046] The air conditioner 1 of the first model without actual operation record and the air conditioner 2 of the second model with actual operation record operate on the same working principle, have the same refrigerant type, and have different discharge volumes set for the compressors 11 and 31 in the refrigeration cycles 10 and 30. Since the working principle of the air conditioner 1 and the air conditioner 2 is the same, it is estimated that the logic of determining the refrigerant circulation amount G based on the compressor frequency Hz and the compressor suction gas specific volume v is similar.
[0047] Specifically, the refrigerant circulation rate G1 [kg / s] of the first refrigeration cycle 10 can be considered to be a function of the compressor frequency Hz [1 / s] and the compressor suction gas specific volume v [m 3 / kg] and exclude volume V1[m 3 ] as a coefficient. In addition, the refrigerant circulation amount G2 [kg / s] of the second refrigeration cycle 30 is expressed as the following formula (2) by using the compressor frequency Hz [1 / s] and the compressor suction gas specific volume v [m 3 / kg] and exclude volume V2[m 3 ] is expressed as a formula with .
[0048] [Number 1]
[0049]
[0050] In the present embodiment, information on the discharge volume V1 set for the first compressor 11 is stored in the first setting information storage unit 211 . In addition, information on the discharge volume V2 set for the second compressor 31 is stored in the second setting information storage unit 411 .
[0051] Here, since there is no operational record for the first refrigeration cycle 10 , an NN model cannot be constructed. Therefore, it is considered that the NN model M constructed for the second refrigeration cycle 30 operating on the same operating principle as the first refrigeration cycle 10 is applied to the operation of the equipment in the first refrigeration cycle 10 .
[0052] In this case, the NN model M is constructed based on the operating characteristics of the air conditioner 2, and therefore cannot be directly used as is for determinations related to the operation of the first refrigeration cycle 10 of the air conditioner 1. Specifically, the first compressor 11 of the first refrigeration cycle 10 and the second compressor 31 of the second refrigeration cycle 30 have different discharge volumes, and therefore, even if the NN model M is used as is, the refrigerant circulation amount G1 of the first refrigeration cycle 10 cannot be properly determined based on the compressor frequency Hz and the compressor suction gas specific volume v.
[0053] Therefore, the first control device 20 performs processing for appropriately determining the refrigerant circulation amount G1 of the first refrigeration cycle 10 using the NN model M by correcting one of the data used for calculation of the refrigerant circulation amount G1.
[0054] Hereinafter, a process in which the first control device 20 determines the refrigerant circulation amount G1 related to the first refrigerating cycle 10 using the NN model M by correcting the compressor frequency Hz which is one of the first item of input data related to the operation of the first refrigerating cycle 10 will be described. Figure 3 It is a flowchart showing the flow of processing executed by the first control device 20 .
[0055] First, the first input unit 22 inputs an instruction for determining the refrigerant circulation amount G1. At this time, the first input unit 22 inputs the information of the compressor frequency Hz and the compressor suction gas specific volume v as the input data of the first item as the explanatory variables provided to the NN model M for determining the refrigerant circulation amount G1. The first input unit 22 also inputs the correction object information indicating that the data of the correction object when the NN model M is used is the compressor frequency Hz, and the information of the discharge volume V2 of the second compressor 31 used for the correction processing of the compressor frequency Hz.
[0056] The information input from the first input unit 22 is acquired by the first information acquisition unit 231. When the first information acquisition unit 231 acquires the instruction to determine the refrigerant circulation amount G1 of the first refrigeration cycle 10 ("Yes" in S1), the first information acquisition unit 231 acquires the input data of the first item acquired together with the instruction, that is, the information of the compressor frequency Hz and the compressor suction gas specific volume v, the information of the discharge volume V2 of the second compressor 31, and the information of the discharge volume V1 of the first compressor 11 stored in the first setting information storage unit 211 as information used for the determination process (S2).
[0057] As described above, it can be predicted that the exclusion volume is a coefficient when determining the refrigerant circulation amount based on the compressor frequency and the compressor suction gas specific volume. That is, it can be considered that in the NN model M, the refrigerant circulation amount is determined by multiplying the input data represented by the compressor frequency and the compressor suction gas specific volume as input data by a value equivalent to the exclusion volume. The first compressor 11 and the second compressor 31 have different exclusion volumes, but in the NN model M, the value equivalent to the exclusion volume is processed in the black box of the intermediate layer, so it cannot be corrected.
[0058] Therefore, based on the correction object information acquired by the first information acquisition unit 231, the first correction unit 232 multiplies the compressor frequency Hz by the ratio of the displacement volume V1 of the first compressor 11 to the displacement volume V2 of the second compressor 31 as shown in the following formula (3), thereby correcting the compressor frequency Hz to the compressor frequency Hz1 (S3).
[0059] [Number 2]
[0060]
[0061] The first determination unit 233 provides the compressor frequency Hz1 corrected by the first correction unit 232 and the compressor suction gas specific volume v obtained by the first information acquisition unit 231 as explanatory variables to the input layer of the neural network. Then, based on the provided explanatory variables, the first determination unit 233 determines the target variable output by the NN model M as the refrigerant circulation amount G1 of the first refrigeration cycle 10 (S4). The NN model M used here corresponds to the second refrigeration cycle 30, so the determined refrigerant circulation amount G1 is expressed as shown in the upper part of the following formula (4) using the information of the discharge volume V2 of the second compressor 31.
[0062] [Number 3]
[0063]
[0064] That is, the first determination unit 233 can determine the refrigerant circulation amount G1 of the first refrigeration cycle 10 corresponding to the discharge volume V1 of the first compressor 11 using the NN model M by using the correction value Hz1 of the compressor frequency.
[0065] The first operation control unit 234 controls the operation of each device 11, 12, 13 or 14 in the first refrigeration cycle 10 based on the refrigerant circulation amount G1 determined by the first determination unit 233 and various setting information stored in the first setting information storage unit 211 (S5).
[0066] According to the first embodiment described above, by correcting input data using the NN model constructed for the second model air conditioner with actual operation records, it is possible to accurately determine information used in the operation of the first model air conditioner with no actual operation records.
[0067] The following case is described: in the above-mentioned determination process, the first CPU 23 corrects the compressor frequency Hz, which is one of the input data of the first item, and determines the refrigerant circulation amount G1 based on the correction value Hz1 of the compressor frequency Hz and the input compressor suction gas specific volume v. However, this is not limited to this, and the compressor suction gas specific volume v, which is one of the input data of the first item, may be corrected in the same manner, and the refrigerant circulation amount G1 may be determined based on the correction value of the compressor suction gas specific volume v and the input compressor frequency Hz.
[0068] <<Second Implementation Method>>
[0069] An air conditioner using a determination device according to a second embodiment of the present invention is described. In the second embodiment, in a newly developed third model air conditioner with no actual operation record, a device in the air conditioner is controlled using an NN model that has been constructed for controlling a second model air conditioner with actual operation record.
[0070] <Configuration of Air Conditioner Using Determination Device According to Second Embodiment>
[0071] Reference Figure 4 The configuration of an air conditioner using the determination device according to the second embodiment of the present invention will be described. The air conditioner 3 according to the present embodiment is a third model air conditioner that has been newly developed and has no operational records.
[0072] The air conditioner 3 includes a third compressor 51, a third evaporator 52, a third electronic expansion valve 53, a third condenser 54, and a third control device 60 as a determination device, which constitute a third refrigeration cycle 50. The functions of the devices 51, 52, 53, and 54 constituting the third refrigeration cycle 50 are the same as the functions of the devices 11, 12, 13, and 14 described in the first embodiment, and therefore detailed descriptions thereof are omitted.
[0073] The third control device 60 includes a third storage unit 61, a third input unit 62, and a third CPU 63, and controls the devices 51, 52, 53, and 54 in the third refrigeration cycle 50. The third storage unit 61 includes a third setting information storage unit 611 and a third NN model information storage unit 612.
[0074] The third setting information storage unit 611 stores setting information related to each device 51, 52, 53, and 54 in the third refrigeration cycle 50. The third NN model information storage unit 612 stores the NN model M described above.
[0075] The third input unit 62 inputs instructions related to the operation of the devices 51, 52, 53, and 54 in the third refrigeration cycle 50 and information used in the processing corresponding to the instructions. Specifically, the third input unit 62 inputs input data of a first item related to the operation of the devices in the third refrigeration cycle 50, and an instruction to output output data of a second item related to the operation of the devices in the third refrigeration cycle 50 from the input data of the first item.
[0076] The third CPU 63 includes a third information acquisition unit 631 , a third determination unit 632 , a third correction unit 633 , and a third operation control unit 634 .
[0077] The third information acquisition unit 631 acquires the command input from the third input unit 62 and the information used in the process corresponding to the command. The third information acquisition unit 631 acquires the information used in the process corresponding to the acquired command from the information stored in the third setting information storage unit 611.
[0078] The third determination unit 632 determines the output data of the second item using the NN model M based on the input data of the first item acquired by the third information acquisition unit 631 .
[0079] The third correction unit 633 multiplies the output data of the second item determined by the third determination unit 632 by the ratio of the first coefficient to the second coefficient, thereby correcting the output data of the corresponding second item to correspond to the third refrigeration cycle 50 .
[0080] The third operation control unit 634 controls the operation of each device 51 , 52 , 53 or 54 in the third refrigeration cycle 50 based on the output data of the second item related to the third refrigeration cycle 50 corrected by the third determination unit 632 and the information stored in the second setting information storage unit 411 .
[0081] <Operation of Air Conditioner Using Determination Device According to Second Embodiment>
[0082] The operation of the air conditioner 3 of this embodiment is explained by taking the following regression model as an example: a regression model in which the compressor frequency Hz used in the third compressor 51 in the third refrigeration cycle 50 and the compressor intake gas specific volume v of the third compressor 51 are used as explanatory variables, and the refrigerant circulation amount G3 in the third refrigeration cycle 50 is used as the target variable.
[0083] The following describes the processing of the third control device 60 for correcting the refrigerant circulation amount G2 output by the NN model M based on the compressor frequency Hz and the compressor suction gas specific volume v related to the operation of the third refrigeration cycle 50, thereby determining the refrigerant circulation amount G3 related to the third refrigeration cycle 50. Figure 5It is a flowchart showing the flow of processing executed by the third control device 60 .
[0084] First, the third input unit 62 inputs an instruction for determining the refrigerant circulation amount G3. At this time, the third input unit 62 inputs the information of the compressor frequency Hz and the compressor suction gas specific volume v as the input data of the first item as the explanatory variables provided to the NN model M for determining the refrigerant circulation amount G3. The third input unit 62 also inputs the correction object information indicating that the correction object data when the NN model M is used is the refrigerant circulation amount G2, and the information of the discharge volume V2 of the second compressor 31 used for the correction processing of the refrigerant circulation amount G2.
[0085] The information input from the third input unit 62 is acquired by the third information acquisition unit 631. When the third information acquisition unit 631 acquires the instruction to determine the refrigerant circulation amount G3 of the third refrigeration cycle 50 ("Yes" in S11), the third information acquisition unit 631 acquires the input data of the first item acquired together with the instruction, that is, the information of the compressor frequency Hz and the compressor suction gas specific volume v, the information of the discharge volume V2 of the second compressor 31, and the information of the discharge volume V3 of the third compressor 51 stored in the third setting information storage unit 611 as information used for the determination process (S12).
[0086] Next, the third determination unit 632 provides the compressor frequency Hz and the compressor suction gas specific volume v acquired by the third information acquisition unit 631 as explanatory variables to the NN model M. Then, the third determination unit 632 determines that the target variable output by the NN model M based on the provided explanatory variables is the refrigerant circulation amount G2 (S13).
[0087] Next, as shown in the following formula (5), the third correction unit corrects the refrigerant circulation volume G2 to the refrigerant circulation volume G3 corresponding to the third refrigeration cycle 50 by multiplying the refrigerant circulation volume G2 determined by the third determination unit 632 by the ratio of the exclusion volume V3 of the third compressor 51 to the exclusion volume V2 of the second compressor 31 (S14).
[0088] [Number 4]
[0089]
[0090] The third operation control unit 634 controls the operation of each device 51, 52, 53 or 54 in the third refrigeration cycle 50 based on the refrigerant circulation amount G3 corrected by the third correction unit 633 and various setting information stored in the third setting information storage unit 611 (S15).
[0091] According to the second embodiment described above, the first type of air conditioner with no operating record can be appropriately controlled by correcting output data using the NN model constructed for the second type of air conditioner with operating record.
[0092] In the first and second embodiments described above, regarding the refrigeration cycle of the air conditioner, when the refrigerant circulation amount is determined using the NN model based on the compressor frequency and the compressor suction gas specific volume, a process of correcting any of the corresponding information is described. However, this is not limited to this, and in other actuators whose workload with respect to a predetermined input parameter differs depending on the model of the equipment, the same technology can be used to determine information related to the work.
[0093] For example, in the refrigeration cycle of the air conditioner, it is conceivable to set at least one of the high pressure, low pressure, exhaust gas temperature, intake gas temperature, inlet water temperature, outside air temperature, water flow rate, compressor frequency or fan speed as input data of the first item, and use the NN model to set the electronic expansion valve opening, fan speed, judgment information on the presence or absence of refrigerant leakage or judgment information on the presence or absence of failure as output data of the second item. In these cases, when the refrigeration cycle operates based on the prescribed input parameters, by correcting any one of the compressor frequency, fan speed or electronic expansion valve opening that varies depending on the model, it is possible to use the NN model related to the existing other models of air conditioners to determine information related to the air conditioner to be controlled.
[0094] [Verification results]
[0095] The following describes the results of verifying the opening degree of the electronic expansion valve under various conditions for a verification target air conditioner, model B, using the NN model constructed for the operation of a model A air conditioner with actual operation performance.
[0096] Figure 6 (a) is a graph showing positions as a set of points, wherein the positions show the corresponding states of the opening of the electronic expansion valve (electronic expansion valve opening prediction value) determined using the NN model associated with type A air conditioner without correcting any data for various operating conditions and the opening of the electronic expansion valve actually measured in type B air conditioner (electronic expansion valve opening actual measurement value).
[0097] Figure 6(b) is a graph showing positions as a set of points, wherein the positions show a state in which, for various operating conditions, any data related to the opening of the electronic expansion valve is corrected to correspond to the air conditioner of type B as described in the above embodiment using the NN model related to the air conditioner of type A, thereby determining a state in which the predicted value of the opening of the electronic expansion valve corresponds to the actual measured value of the opening of the electronic expansion valve actually measured in the air conditioner of type B under the corresponding conditions.
[0098] exist Figure 6 In (a) and (b), when the predicted value of the electronic expansion valve opening is consistent with the actual value of the electronic expansion valve opening under predetermined operating conditions, a point corresponding to an arbitrary position on the dotted line L in the graph is shown.
[0099] exist Figure 6 In (a), the set of points deviates greatly from the dotted line L. In contrast, Figure 6 In (b), the set of points is concentrated at a position close to the dotted line L. Thus, it was verified that by correcting any data related to the opening of the electronic expansion valve, the opening of the electronic expansion valve can be determined with high accuracy using the NN model related to the A-type air conditioner for the B-type air conditioner.
[0100] The embodiments of the present invention have been described, but the embodiments are presented as examples and are not intended to limit the scope of the invention. The new embodiments can be implemented in various other ways, and various omissions, substitutions, and changes can be made without departing from the scope of the subject matter of the invention. The embodiments and their variations are included in the scope and subject matter of the invention, and are included in the invention described in the claims and the scope equivalent thereto.
[0101] Furthermore, by programming the functional configuration of the first control device 20 or the third control device 60 and installing it in a computer, it is also possible to construct a determination program that causes the computer to function as the first control device 20 or the third control device 60 .
Claims
1. A determination device, characterized in that: have: A model information storage unit storing a neural network model constructed for the operation of a second device that operates on the same working principle as the first device as the control target and is a different model; an information acquisition unit, which acquires input data of a first item related to a device operating according to the working principle, a first coefficient for calculating output data of a second item based on the input data of the first item in the first device, and a second coefficient for calculating output data of the second item based on the input data of the first item in the second device; a determination unit that determines the output data of the second item related to the first device based on the input data of the first item using the neural network model stored in the model information storage unit; The correction unit corrects the corresponding data by multiplying the input data of the first item or the output data of the second item used in the determination unit by the ratio of the first coefficient to the second coefficient.
2. The determination device according to claim 1, characterized in that: The correction unit corrects data related to a brake, of the input data of the first item or the output data of the second item used by the determination unit, the amount of work of the brake differing according to a model of a device with respect to a predetermined input parameter.
3. The determination device according to claim 1 or 2, characterized in that: The information acquisition unit uses at least one of high pressure, low pressure, exhaust gas temperature, intake gas temperature, inlet water temperature, outside air temperature, water flow rate, compressor frequency or fan speed as input data of the first item, and uses the electronic expansion valve opening, fan speed, judgment information on whether there is refrigerant leakage or whether there is a fault as output data of the second item to acquire information, The correction unit corrects any one of a compressor frequency, a fan rotation speed, or an opening degree of an electronic expansion valve.
4. A determination method, characterized in that: storing a neural network model constructed for the operation of a second device that operates on the same working principle as the first device as the control target and is a different model, obtaining input data of a first item related to a device operating according to the working principle, a first coefficient for calculating output data of a second item based on the input data of the first item in the first device, and a second coefficient for calculating output data of the second item based on the input data of the first item in the second device, The input data of the first item is multiplied by the ratio of the first coefficient to the second coefficient for correction, and the output data of the second item related to the first device is determined using the neural network model based on the corrected input data of the first item, or the output data of the second item is determined using the neural network model based on the input data of the first item, and then the output data of the determined second item is multiplied by the ratio of the first coefficient to the second coefficient to correct it in a manner corresponding to the first device.
5. A determination program, characterized in that: Enable the computer to perform the following functions: storing a neural network model constructed for the operation of a second device that operates on the same working principle as the first device as the control target and is a different model; Acquire input data of a first item related to a device operating according to the working principle, a first coefficient for calculating output data of a second item based on the input data of the first item in the first device, and a second coefficient for calculating output data of the second item based on the input data of the first item in the second device; The input data of the first item is multiplied by the ratio of the first coefficient to the second coefficient for correction, and the output data of the second item related to the first device is determined using the neural network model based on the corrected input data of the first item, or the output data of the second item is determined using the neural network model based on the input data of the first item, and then the output data of the determined second item is multiplied by the ratio of the first coefficient to the second coefficient to correct it in a manner corresponding to the first device.