Refrigerant leakage detection device and method

Abstract

The present invention generalizes a prediction model of a refrigerant amount index value. One embodiment of the present disclosure pertains to a refrigerant leakage detection device (10) for detecting refrigerant leakage in a refrigeration / air-conditioning apparatus provided with a refrigeration cycle in which a refrigerant circulates through a refrigerant circuit in which a compressor, a condenser, an expansion valve, and an evaporator are connected by piping. A control unit (1001) of the refrigerant leakage detection device inputs steady-state data determined for the refrigeration / air-conditioning apparatus and operation data of the refrigeration / air-conditioning apparatus to a prediction model (1000), predicts a refrigerant amount index value, and detects refrigerant leakage in the refrigerant circuit using the predicted refrigerant amount index value.

Description

Refrigerant leak detection device and method

[0001] FIELD OF THE DISCLOSURE The present disclosure relates to refrigerant leak detection devices and methods.

[0002] Conventionally, it has been known to use a prediction model to predict values ​​that serve as indicators of the refrigerant amount of a refrigeration and air conditioning equipment (hereinafter also referred to as refrigerant amount index values, for example, expansion valve opening, subcooling degree, intake superheat degree, discharge superheat degree, etc.) from the operating data of the refrigeration and air conditioning equipment, and to detect refrigerant leaks in the refrigeration and air conditioning equipment.

[0003] Patent No. 6410990

[0004] Even if the configuration of the refrigerant circuit through which the refrigerant circulates in a refrigeration / air conditioning equipment is the same, the balance point of the refrigeration cycle will differ if the performance or operating conditions of the refrigeration / air conditioning equipment differ. Therefore, a prediction model for the refrigerant amount index value must be prepared for each performance or operating condition of the refrigeration / air conditioning equipment.

[0005] The present disclosure aims to generalize a prediction model so that the same refrigerant amount index value prediction model can be used even when the performance and usage conditions of the refrigeration and air conditioning equipment differ.

[0006] A refrigerant leak detection device according to a first aspect of the present disclosure is a refrigerant leak detection device (10) that detects refrigerant leaks in refrigeration and air conditioning equipment having a refrigeration cycle in which refrigerant circulates in a refrigerant circuit in which a compressor, a condenser, an expansion valve, and an evaporator are connected by piping, and a control unit (1001) of the refrigerant leak detection device inputs steady-state data determined for each refrigeration and air conditioning equipment and operating data of the refrigeration and air conditioning equipment into a prediction model (1000) to predict a refrigerant quantity index value, and detects refrigerant leaks in the refrigerant circuit using the predicted refrigerant quantity index value.

[0007] According to the first aspect of the present disclosure, since the difference in the balance point of the refrigeration cycle caused by various differences between refrigeration and air conditioning equipment can be corrected by using steady-state data determined for each refrigeration and air conditioning equipment, the prediction model for the refrigerant amount index value can be generalized, and as a result, the same prediction model can be used for refrigeration and air conditioning equipment with different attributes such as performance and operating conditions.

[0008] A second aspect of the present disclosure is the refrigerant leak detection device according to the first aspect, wherein the steady-state data determined for each refrigeration and air conditioning equipment includes data indicating the performance of the refrigeration and air conditioning equipment and data indicating the usage conditions of the refrigeration and air conditioning equipment.

[0009] According to the second aspect of the present disclosure, it is possible to correct differences in the balance point of the refrigeration cycle caused by differences in the performance of the refrigeration and air conditioning equipment, or differences in the balance point of the refrigeration cycle caused by differences in the usage conditions of the refrigeration and air conditioning equipment.

[0010] A third aspect of the present disclosure is a refrigerant leak detection device according to the first or second aspect, wherein the prediction model is generated by machine learning using learning data in which operating data of a plurality of refrigeration and air conditioning equipment having different performance or usage conditions and data indicating the performance or usage conditions are input, and a refrigerant quantity index value is output.

[0011] According to the third aspect of the present disclosure, it is possible to correct differences in the balance point of the refrigeration cycle caused by differences in the performance of the refrigeration and air conditioning equipment, or differences in the balance point of the refrigeration cycle caused by differences in the usage conditions of the refrigeration and air conditioning equipment.

[0012] A fourth aspect of the present disclosure is the refrigerant leak detection device according to the second or third aspect, wherein the performance of the refrigeration and air conditioning equipment is a capacity of the refrigeration and air conditioning equipment.

[0013] According to the fourth aspect of the present disclosure, it is possible to correct a difference in the balance point of the refrigeration cycle caused by a difference in equipment capacity, which is the refrigeration or air conditioning capacity exhibited by the refrigeration and air conditioning equipment.

[0014] A fifth aspect of the present disclosure is the refrigerant leak detection device according to any one of the first to fourth aspects, wherein the steady-state data determined for each refrigeration and air conditioning device is expressed as qualitative data.

[0015] According to the fifth aspect of the present disclosure, it is possible to correct differences in the balance point of the refrigeration cycle caused by differences in refrigeration and air conditioning equipment represented by qualitative data.

[0016] A sixth aspect of the present disclosure is the refrigerant leakage detection device according to any one of the first to fourth aspects, wherein the steady-state data determined for each refrigeration and air conditioning device is expressed as quantitative data.

[0017] According to the sixth aspect of the present disclosure, it is possible to correct differences in the balance point of the refrigeration cycle caused by differences in refrigeration and air conditioning equipment represented by quantitative data.

[0018] A method according to a seventh aspect of the present disclosure is a method performed by a refrigerant leak detection device (10) that detects refrigerant leaks in refrigeration and air conditioning equipment having a refrigeration cycle in which refrigerant circulates in a refrigerant circuit in which a compressor, a condenser, an expansion valve, and an evaporator are connected by piping, and the method includes inputting steady-state data determined for each refrigeration and air conditioning equipment and operating data of the refrigeration and air conditioning equipment into a prediction model (1000) to predict a refrigerant quantity index value, and detecting refrigerant leaks in the refrigerant circuit using the predicted refrigerant quantity index value.

[0019] FIG. 1 is a diagram for explaining an overview of the present disclosure. FIG. 2 is a diagram for explaining an overview of the present disclosure. FIG. 3 is a diagram showing an overall configuration according to an embodiment of the present disclosure. FIG. 4 is a diagram showing hardware configurations of a refrigerant leak detection device and a learning device according to an embodiment of the present disclosure. FIG. 5 is a diagram showing a functional configuration of a refrigerant leak detection device according to an embodiment of the present disclosure. FIG. 6 is a diagram showing a functional configuration of a learning device according to an embodiment of the present disclosure. FIG. 7 is a flowchart of a refrigerant leak detection process according to an embodiment of the present disclosure. FIG. 8 is a flowchart of a learning process according to an embodiment of the present disclosure.

[0020] Hereinafter, embodiments of the present disclosure will be described with reference to the drawings.

[0021] <Explanation of Terms> In the present disclosure, a refrigerant quantity index value can be predicted from data related to refrigeration and air conditioning equipment and operational data of the refrigeration and air conditioning equipment. "Data related to refrigeration and air conditioning equipment" is steady-state data determined for each refrigeration and air conditioning equipment. For example, data related to refrigeration and air conditioning equipment includes at least one of data indicating the performance of the refrigeration and air conditioning equipment and data indicating the operating conditions of the refrigeration and air conditioning equipment. Note that the data related to refrigeration and air conditioning equipment may be attributes of the refrigeration and air conditioning equipment or constants that do not change during operation of the refrigeration and air conditioning equipment. "Operational data of refrigeration and air conditioning equipment" is one or more variables (i.e., non-steady-state data) acquired by various sensors during operation of the refrigeration and air conditioning equipment.

[0022] <Overview> First, an overview of the present disclosure will be described with reference to Figures 1 and 2. As shown in Figure 1, it is assumed that operating data of a refrigeration and air conditioning device having a refrigeration cycle in which a refrigerant circulates in a refrigerant circuit in which a compressor 101, a condenser 102, an expansion valve 103, and an evaporator 104 are connected by piping is input to a prediction model, and a refrigerant amount index value is output.

[0023] When refrigeration and air conditioning equipment specifications differ, the piping, heat exchangers, compressors, and other components that make up the refrigerant circuit are selected to be appropriately sized to match the equipment capacity. For example, when designing Model B, which has twice the equipment capacity of Model A, it is theoretically possible to calculate the pipe diameter that will result in the same pressure loss as Model A's piping, and the heat exchangers and compressors that will have twice the capacity of Model A's specifications. However, when actually commercializing the product, the piping is selected from existing standard products, so the piping pressure loss values ​​for Models A and B will be similar, but not identical. Similarly, heat exchangers and compressors face various constraints, such as material, design, and cost, making it difficult to precisely double the capacity of the finished product. In the case of compressors, even if the capacity can be doubled, the mechanical and thermal loss characteristics vary depending on the equipment size, resulting in different compressor efficiencies. Therefore, even if the refrigeration circuit configuration of a refrigeration and air conditioning equipment is the same, different specifications will generally result in different balance points (vertices on the pH diagram) for the refrigeration cycle, even when operating under the same environmental conditions (same outdoor temperature, outdoor humidity, indoor temperature, indoor humidity, and ratio of equipment capacity to load).

[0024] For example, as shown in the ph diagram in Figure 2, even if model A and model B are operated under the same environmental conditions, the balance points are different, and the shapes of the refrigeration cycles on the diagram are different. Therefore, separate prediction models for model A and model B must be prepared.

[0025] Furthermore, even if the refrigerant circuit configuration is the same and the specifications of the refrigeration and air conditioning equipment are the same, if the operating conditions, such as the number of years since the installation of the refrigeration and air conditioning equipment, are different, the balance point of the refrigeration cycle may fluctuate due to deterioration of the equipment's components over time, even when the equipment is operated under the same conditions (same outdoor temperature, outdoor humidity, indoor temperature, indoor humidity, and ratio of equipment capacity to load).

[0026] As such, the balance point (each vertex of the pH diagram) of the refrigeration cycle differs depending on the performance of the refrigeration and air conditioning equipment or the operating conditions of the refrigeration and air conditioning equipment. Therefore, a prediction model for the refrigerant quantity index value must be generated for each performance or operating condition of the refrigeration and air conditioning equipment. In the present disclosure, a prediction model is generated by machine learning using training data that includes numerical data (quantitative data) or classification labels (qualitative data) indicating the performance or operating conditions of multiple refrigeration and air conditioning equipment with different performance or operating conditions. This allows for the generation of a general-purpose model (common model) that can be used with refrigeration and air conditioning equipment with different performance or operating conditions. Therefore, even if sufficient training data cannot be prepared for a single model, a prediction model that can be used across multiple models can be generated.

[0027] <Overall Configuration> Fig. 3 is a diagram showing the overall configuration according to one embodiment of the present disclosure. The refrigerant leak detection system 1 can include a refrigerant leak detection device 10 and a learning device 20. Note that, although Fig. 3 shows the refrigerant leak detection device 10 and the learning device 20 as separate devices, the refrigerant leak detection device 10 and the learning device 20 may also be combined into a single device. Each will be described below.

[0028] <<Refrigerant Leak Detection System>> The refrigerant leak detection system 1 is a system that detects refrigerant leakage from the refrigeration and air conditioning equipment 30.

[0029] <<Refrigerant Leak Detection Device>> The refrigerant leak detection device 10 inputs data related to the refrigeration and air conditioning equipment 30 and operating data of the refrigeration and air conditioning equipment 30 into the prediction model generated by the learning device 20, outputs a refrigerant amount index value, and uses the refrigerant amount index value to detect a refrigerant leak in the refrigeration and air conditioning equipment 30. The refrigerant leak detection device 10 is composed of one or more computers.

[0030] <<Learning Device>> The learning device 20 is a device that generates a prediction model. The explanatory variables of the prediction model are data related to the refrigeration and air conditioning equipment 30 and operating data of the refrigeration and air conditioning equipment 30, and the objective variable of the prediction model is a refrigerant amount index value. The learning device 20 is composed of one or more computers.

[0031] <<Refrigeration and Air Conditioning Equipment>> The refrigeration and air conditioning equipment 30 includes a refrigeration cycle in which a refrigerant circulates through a refrigerant circuit including a compressor, a condenser, an expansion valve, and an evaporator connected by piping. The refrigeration cycle may further include a subcooling heat exchanger, an economizer, and an expansion valve for controlling the capacity of the subcooling heat exchanger or the economizer. The refrigeration and air conditioning equipment 30 may be a direct expansion type or an indirect expansion type, and may be used to cool air or a secondary heat medium (e.g., water) (e.g., during cooling) or to heat air or a secondary heat medium (e.g., water) (e.g., during heating).

[0032] Here, the "data relating to refrigeration and air conditioning equipment," "operation data of refrigeration and air conditioning equipment," and "refrigerant amount index value of refrigeration and air conditioning equipment" used in this disclosure will be described.

[0033] [Data Related to Refrigeration and Air Conditioning Equipment] Data related to the refrigeration and air conditioning equipment 30 is data that is a factor that causes differences in the balance points (vertices of the pH diagram) of the refrigeration cycle other than environmental conditions such as the outside air temperature, room temperature, indoor heat generation load, etc. For example, the data related to the refrigeration and air conditioning equipment 30 includes data indicating the performance of the refrigeration and air conditioning equipment 30 and data indicating the usage conditions of the refrigeration and air conditioning equipment 30.

[0034] [Data Indicating the Performance of Refrigeration and Air Conditioning Equipment] For example, data indicating the performance of refrigeration and air conditioning equipment is equipment capacity. For example, equipment capacity may be a refrigeration capacity or an air conditioning capacity (for example, kW, HP, or refrigeration tons), a size of a room to be air-conditioned (for example, m ２ For example, equipment capacity is expressed as a qualitative variable such as an identification number (for example, 1, 2, 3, ...) corresponding to the difference in equipment capacity.

[0035] For example, data indicating the performance of the refrigeration and air conditioning equipment is the generation (hereinafter also referred to as equipment generation) of the refrigeration and air conditioning equipment 30. For example, the equipment generation is expressed by a qualitative variable such as current, one generation before, or two generations before.

[0036] For example, data indicating the performance of the refrigeration and air conditioning equipment is the specifications (hereinafter also referred to as equipment specifications) of the refrigeration and air conditioning equipment 30. For example, the equipment specifications are expressed by qualitative variables indicating the type of the refrigeration and air conditioning equipment 30 based on the energy consumption efficiency, such as an energy-saving type or a standard type.

[0037] For example, data indicating the performance of the refrigeration and air conditioning equipment is the specifications of the components (hereinafter also referred to as component specifications) used by the refrigeration and air conditioning equipment 30. For example, component specifications are specifications of a compressor, an expansion valve, a heat exchanger, a refrigerant communication pipe length, etc. For example, component specifications may be expressed by quantitative variables such as the capacity (e.g., kW) of the compressor and the expansion valve, and the size (e.g., m) of the refrigerant communication pipe length, or by qualitative variables such as identification numbers (e.g., 1, 2, 3, ...) corresponding to differences in capacity or size.

[0038] For example, data indicating the performance of the refrigeration and air conditioning equipment is the specifications of the refrigerant (hereinafter also referred to as refrigerant specifications) circulating through the refrigerant circuit of the refrigeration and air conditioning equipment 30. For example, the refrigerant specifications are expressed by qualitative variables indicating the type of refrigerant, such as R410A, R32, R134a, etc.

[0039] [Data Indicating Usage Conditions of the Refrigeration and Air Conditioning Equipment] For example, the data indicating the usage conditions of the refrigeration and air conditioning equipment is a period such as the number of years that the refrigeration and air conditioning equipment 30 has been in use (hereinafter also referred to as years of use). By using the number of years of use of the refrigeration and air conditioning equipment 30, it is possible to correct differences in the balance point of the refrigeration cycle due to differences in deterioration over time. For example, the number of years of use may be expressed as a quantitative variable (e.g., a numerical value such as 5 years or 10 years) or as a qualitative variable (e.g., a rank such as less than 5 years, 5 to 10 years, or more than 10 years).

[0040] For example, data indicating the usage conditions of the refrigeration and air conditioning equipment is the purpose of the installed refrigeration and air conditioning equipment 30 (hereinafter also referred to as the installation purpose). By using the installation purpose of the refrigeration and air conditioning equipment 30, it is possible to correct differences in the balance point of the refrigeration cycle due to differences in load patterns. For example, the installation purpose is expressed by a qualitative variable such as office, hospital, school, server room, etc.

[0041] For example, the data indicating the usage conditions of the refrigeration and air conditioning equipment is the region (e.g., a country, a region consisting of multiple countries, a region within a country, etc., hereinafter also referred to as the installation region) in which the refrigeration and air conditioning equipment 30 is installed. By using the installation region of the refrigeration and air conditioning equipment 30, it is possible to correct for differences in the balance point of the refrigeration cycle due to differences in weather conditions and load patterns. For example, the installation region is expressed by a qualitative variable such as Japan, North America, Europe, India, etc.

[0042] [Operational Data of Refrigeration and Air Conditioning Equipment] The operational data of the refrigeration and air conditioning equipment 30 is one or more variables that can be acquired during operation of the refrigeration and air conditioning equipment 30. For example, the operational data of the refrigeration and air conditioning equipment 30 includes at least one of the following: outdoor air temperature, supply water temperature, return water temperature, supply water temperature set value, pump rotation speed (amount of water circulation), compressor load factor, inverter compressor rotation speed, number of operating constant speed compressors, degree of subcooling, intake superheat, target intake superheat value, discharge superheat, condensation temperature, target condensation temperature value, evaporation temperature, target evaporation temperature value, expansion valve opening, outdoor unit fan rotation speed, indoor unit fan step

[0043] [Refrigerant Amount Index Value of Refrigeration and Air Conditioning Equipment] At least one of the expansion valve opening, the degree of subcooling, the degree of intake superheat, the degree of discharge superheat, and a calculated value calculated using any one or a combination of the expansion valve opening, the degree of subcooling, the degree of intake superheat, and the degree of discharge superheat can be used as the refrigerant amount index value of the refrigeration and air conditioning equipment 30. Furthermore, if the refrigeration and air conditioning equipment 30 is equipped with a subcooling heat exchanger or an economizer, the degree of subcooling at the outlet of the subcooling heat exchanger or the economizer, and the expansion valve opening for controlling the subcooling heat exchanger or the economizer may be used in addition to the above refrigerant amount index.

[0044] The expansion valve opening (EV opening) is the opening of the expansion valve (EV) 103 .

[0045] The degree of subcooling is the difference between the condenser outlet temperature Tb and the condensation temperature Tc.

[0046] The degree of suction superheat is the difference between the compressor suction temperature Ts and the evaporation temperature Te.

[0047] The discharge superheat degree is the difference between the compressor discharge temperature Td and the condensation temperature Tc.

[0048] 4 is a diagram showing the hardware configuration of the refrigerant leak detection device 10 and the learning device 20 according to an embodiment of the present disclosure. The refrigerant leak detection device 10 and the learning device 20 can include a control unit 1001, a main memory unit 1002, an auxiliary memory unit 1003, an input unit 1004, an output unit 1005, and an interface unit 1006. Each of these will be described below.

[0049] The control unit 1001 is a processor (for example, a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), etc.) that executes various programs installed in the auxiliary storage unit 1003 .

[0050] The main memory unit 1002 includes a non-volatile memory (Read Only Memory (ROM)) and a volatile memory (Random Access Memory (RAM)). The ROM stores various programs, data, etc. required for the control unit 1001 to execute various programs installed in the auxiliary memory unit 1003. The RAM provides a working area into which the various programs installed in the auxiliary memory unit 1003 are expanded when executed by the control unit 1001.

[0051] The auxiliary storage unit 1003 is an auxiliary storage device that stores various programs and information used when the various programs are executed.

[0052] The input unit 1004 is an input device through which the operator of the refrigerant leak detection device 10 and the learning device 20 inputs various instructions to the refrigerant leak detection device 10 and the learning device 20 .

[0053] The output unit 1005 is an output device that outputs the internal states of the refrigerant leak detection device 10 and the learning device 20, etc.

[0054] The interface unit 1006 is a communication device for connecting to a network and communicating with other devices.

[0055] <Functional Configuration> Figure 5 is a diagram showing the functional configuration of the refrigerant leak detection device 10 according to an embodiment of the present disclosure. The control unit 1001 of the refrigerant leak detection device 10 includes a data acquisition unit 111, a prediction unit 112, a calculation unit 113, and a refrigerant leak detection unit 114. Furthermore, the control unit 1001 of the refrigerant leak detection device 10 executes a program to function as the data acquisition unit 111, the prediction unit 112, the calculation unit 113, and the refrigerant leak detection unit 114. The refrigerant leak detection device 10 stores a prediction model 1000 generated by the learning device 20.

[0056] The data acquisition unit 111 acquires data related to refrigeration and air conditioning equipment 30 in operation (i.e., refrigeration and air conditioning equipment that is the target of refrigerant leak detection) from a data server 40 installed remotely from the refrigeration and air conditioning equipment 30, and acquires operating data of the refrigeration and air conditioning equipment 30 from the refrigeration and air conditioning equipment 30. Note that the data related to the refrigeration and air conditioning equipment 30 may be acquired from the refrigeration and air conditioning equipment 30 itself or from an external controller installed near the refrigeration and air conditioning equipment 30 to manage multiple installed refrigeration and air conditioning equipment 30.

[0057] The prediction unit 112 uses the prediction model 1000 to predict a refrigerant amount index value from the data related to the refrigeration and air conditioning equipment 30 acquired by the data acquisition unit 111 and the operating data of the refrigeration and air conditioning equipment 30. Specifically, the prediction unit 112 inputs the data related to the refrigeration and air conditioning equipment 30 acquired by the data acquisition unit 111 and the operating data of the refrigeration and air conditioning equipment 30 into the prediction model 1000, and outputs a refrigerant amount index value.

[0058] The calculation unit 113 calculates a refrigerant amount index value (i.e., an actually measured value). Specifically, the calculation unit 113 calculates the refrigerant amount index value using the operating data acquired by the data acquisition unit 111.

[0059] The refrigerant leak detection unit 114 detects a refrigerant leak. For example, the refrigerant leak detection unit 114 determines that a refrigerant leak has occurred if the difference between the refrigerant amount index value predicted by the prediction unit 112 and the refrigerant amount index value calculated by the calculation unit 113 is equal to or greater than a threshold, and determines that a refrigerant leak has not occurred if the difference between the refrigerant amount index value predicted by the prediction unit 112 and the refrigerant amount index value calculated by the calculation unit 113 is less than the threshold. If a refrigerant leak has occurred, the refrigerant leak detection unit 114 can notify any device that a refrigerant leak has occurred.

[0060] The prediction model 1000 is a model in which explanatory variables are data related to the refrigeration and air conditioning equipment 30 and operating data of the refrigeration and air conditioning equipment 30, and the objective variable is the refrigerant amount index value.

[0061] 6 is a diagram showing the functional configuration of a learning device 20 according to an embodiment of the present disclosure. A control unit 2001 of the learning device 20 includes a teacher data acquisition unit 211 and a learning unit 212. The control unit 2001 of the learning device 20 also functions as the teacher data acquisition unit 211 and the learning unit 212 by executing a program.

[0062] The training data acquisition unit 211 acquires training data used when generating the prediction model 1000 .

[0063] For example, the teacher data is data related to the refrigeration and air conditioning equipment 30, operating data of the refrigeration and air conditioning equipment 30, and a refrigerant amount index value when the refrigeration and air conditioning equipment 30 is operating normally.

[0064] The learning unit 212 generates the prediction model 1000 using the teacher data acquired by the teacher data acquisition unit 211. For example, the learning unit 212 performs machine learning to generate the prediction model 1000 so that the refrigerant amount index value output when the data related to the refrigeration and air conditioning equipment 30, which is the teacher data, and the operation data of the refrigeration and air conditioning equipment 30 are input to the prediction model 1000 approaches the refrigerant amount index value, which is the teacher data.

[0065] For example, the prediction model 1000 is a model that, when data related to the refrigeration and air conditioning equipment 30 and operating data of the refrigeration and air conditioning equipment 30 are input, outputs a refrigerant amount index value when the refrigeration and air conditioning equipment 30 is normal.

[0066] <Processing Method> FIG. 7 is a flowchart of a refrigerant leak detection process according to an embodiment of the present disclosure.

[0067] In step 101 ( S101 ), the control unit 1001 of the refrigerant leakage detection device 10 acquires data related to the refrigeration and air conditioning equipment 30 and operation data of the refrigeration and air conditioning equipment 30 from the refrigeration and air conditioning equipment 30 .

[0068] In step 102 (S102), the control unit 1001 of the refrigerant leak detection device 10 inputs the data regarding the refrigeration and air conditioning equipment 30 acquired in S101 and the operating data of the refrigeration and air conditioning equipment 30 into the prediction model 1000, and outputs a refrigerant quantity index value.

[0069] In step 103 (S103), the control unit 1001 of the refrigerant leakage detection device 10 calculates a refrigerant amount index value (for example, an actually measured value).

[0070] In step 104 (S104), the control unit 1001 of the refrigerant leak detection device 10 detects a refrigerant leak. For example, the control unit 1001 of the refrigerant leak detection device 10 determines that a refrigerant leak has occurred if the difference between the refrigerant amount index value of S102 and the refrigerant amount index value of S103 is equal to or greater than a threshold value, and determines that a refrigerant leak has not occurred if the difference between the refrigerant amount index value of S102 and the refrigerant amount index value of S103 is less than the threshold value.

[0071] FIG. 8 is a flowchart of a learning process according to an embodiment of the present disclosure.

[0072] In step 201 (S201), the control unit 2001 of the learning device 20 acquires training data to be used when generating the prediction model 1000.

[0073] In step 202 (S202), the control unit 2001 of the learning device 20 generates the prediction model 1000 using the training data acquired in S201. For example, the control unit 2001 of the learning device 20 performs machine learning to generate the prediction model 1000 so that the refrigerant amount index value output when the training data related to the refrigeration and air conditioning equipment 30 and the operating data of the refrigeration and air conditioning equipment 30 are input to the prediction model 1000 approaches the refrigerant amount index value that is the training data.

[0074] As described above, according to the embodiment of the present disclosure, there is provided a refrigerant leak detection device (10) that detects refrigerant leaks in refrigeration and air conditioning equipment having a refrigeration cycle in which refrigerant circulates in a refrigerant circuit in which a compressor, a condenser, an expansion valve, and an evaporator are connected by piping, and a control unit (1001) of the refrigerant leak detection device inputs data related to the refrigeration and air conditioning equipment and operating data of the refrigeration and air conditioning equipment into a prediction model (1000) to predict a refrigerant quantity index value, and a refrigerant leak detection device that detects refrigerant leaks in the refrigerant circuit using the predicted refrigerant quantity index value can be provided.

[0075] This allows the difference in the balance point of the refrigeration cycle to be corrected using data related to the refrigeration and air conditioning equipment, making it possible to generalize the prediction model for the refrigerant amount index value.

[0076] Preferably, the data related to the refrigeration and air conditioning equipment includes data indicating the performance of the refrigeration and air conditioning equipment and data indicating the usage conditions of the refrigeration and air conditioning equipment, thereby making it possible to correct a difference in the balance point of the refrigeration cycle caused by a difference in the performance of the refrigeration and air conditioning equipment or a difference in the balance point of the refrigeration cycle caused by a difference in the usage conditions of the refrigeration and air conditioning equipment.

[0077] Preferably, the prediction model is generated by machine learning using learning data in which the inputs are operating data of a plurality of refrigeration and air conditioning devices with different performance or usage conditions and data indicating the performance or usage conditions, and the output is a refrigerant amount index value. This makes it possible to correct differences in the balance point of the refrigeration cycle caused by differences in the performance of the refrigeration and air conditioning devices or differences in the usage conditions of the refrigeration and air conditioning devices.

[0078] Preferably, the performance of the refrigeration and air conditioning equipment is the equipment capacity of the refrigeration and air conditioning equipment, thereby making it possible to correct differences in the balance point of the refrigeration cycle caused by differences in the equipment capacity of the refrigeration and air conditioning equipment.

[0079] Preferably, the performance of the refrigeration and air conditioning equipment is the refrigeration capacity or the air conditioning capacity of the refrigeration and air conditioning equipment, thereby making it possible to correct a difference in the balance point of the refrigeration cycle caused by a difference in the refrigeration capacity or the air conditioning capacity of the refrigeration and air conditioning equipment.

[0080] Preferably, the data relating to the refrigeration and air conditioning equipment is expressed as qualitative data, which makes it possible to correct differences in the balance point of the refrigeration cycle caused by differences in the refrigeration and air conditioning equipment expressed as qualitative data.

[0081] Preferably, the data relating to the refrigeration and air conditioning equipment is expressed as quantitative data, which makes it possible to correct differences in the balance point of the refrigeration cycle caused by differences in the refrigeration and air conditioning equipment expressed as quantitative data.

[0082] According to an embodiment of the present disclosure, a method can be provided that is executed by a refrigerant leak detection device (10) that detects refrigerant leaks in refrigeration and air conditioning equipment having a refrigeration cycle in which refrigerant circulates in a refrigerant circuit in which a compressor, a condenser, an expansion valve, and an evaporator are connected by piping, the method inputting data related to the refrigeration and air conditioning equipment and operating data of the refrigeration and air conditioning equipment into a prediction model (1000), predicting a refrigerant quantity index value, and detecting refrigerant leaks in the refrigerant circuit using the predicted refrigerant quantity index value.

[0083] Although the embodiments have been described above, it will be understood that various changes in form and details can be made without departing from the spirit and scope of the claims.

[0084] This international application claims priority based on Japanese Patent Application No. 2024-146609, filed on August 28, 2024, the entire contents of which are incorporated herein by reference.

[0085] REFRIGERATION SYSTEM 10 REFRIGERATOR LEAK DETECTION DEVICE 20 LEARNING DEVICE 30 REFRIGERATOR AND AIR-CONDITIONING EQUIPMENT 40 DATA SERVER 101 COMPRESSOR 102 CONDENSER 103 EXPANSION VALVE 104 EVAPORATOR 111 DATA ACQUISITION UNIT 112 PREDICTION UNIT 113 CALCULATOR 114 REFRIGERATOR LEAK DETECTION UNIT 211 TEACHER DATA ACQUISITION UNIT 212 LEARNING UNIT 1000 PREDICTION MODEL 1001 CONTROL UNIT 1002 MAIN MEMORY UNIT 1003 AUXILIARY MEMORY UNIT 1004 INPUT UNIT 1005 OUTPUT UNIT 1006 INTERFACE UNIT

Claims

1. A refrigerant leak detection device (10) for detecting refrigerant leaks in refrigeration and air conditioning equipment having a refrigeration cycle in which refrigerant circulates in a refrigerant circuit in which a compressor, a condenser, an expansion valve, and an evaporator are connected by piping, wherein a control unit (1001) of the refrigerant leak detection device inputs steady-state data determined for each refrigeration and air conditioning equipment and operating data of the refrigeration and air conditioning equipment into a prediction model (1000) to predict a refrigerant quantity index value, and detects refrigerant leaks in the refrigerant circuit using the predicted refrigerant quantity index value.

2. A refrigerant leak detection device as described in claim 1, wherein the steady-state data determined for each refrigeration and air conditioning equipment includes data indicating the performance of the refrigeration and air conditioning equipment and data indicating the operating conditions of the refrigeration and air conditioning equipment.

3. A refrigerant leak detection device as described in claim 1 or 2, wherein the predictive model is generated by machine learning using learning data in which the input is operating data of multiple refrigeration and air conditioning equipment with different performance or usage conditions and data indicating the performance or usage conditions, and the output is a refrigerant quantity index value.

4. A refrigerant leak detection device according to claim 2 or 3, wherein the performance of the refrigeration and air conditioning equipment is the equipment capacity of the refrigeration and air conditioning equipment.

5. A refrigerant leak detection device according to any one of claims 1 to 4, wherein the steady-state data determined for each refrigeration and air conditioning device is expressed as qualitative data.

6. A refrigerant leak detection device according to any one of claims 1 to 4, wherein the steady-state data determined for each refrigeration and air conditioning device is expressed as quantitative data.

7. A method executed by a refrigerant leak detection device (10) for detecting refrigerant leaks in refrigeration and air conditioning equipment having a refrigeration cycle in which refrigerant circulates in a refrigerant circuit in which a compressor, a condenser, an expansion valve, and an evaporator are connected by piping, the method comprising: inputting steady-state data determined for each refrigeration and air conditioning equipment and operating data of the refrigeration and air conditioning equipment into a prediction model (1000) to predict a refrigerant quantity index value; and detecting refrigerant leaks in the refrigerant circuit using the predicted refrigerant quantity index value.

Images