Prediction device, refrigeration system, prediction method, and prediction program

Abstract

A defrosting operation can be started at a timing corresponding to frosting. Provided is a prediction device that outputs information for controlling the start timing when a defrosting operation is started by a refrigerator having a refrigerant circuit in which a compressor, a radiator, an expansion valve, and the heat absorber are annularly connected, said defrosting operation being performed on a heat exchanger that functions as the heat absorber. When a learning completion model is generated on the basis of learning data, a control unit of the prediction device inputs current operation data in the refrigerator to the learning completion model, and performs prediction on the current operation data in the refrigerator. The learning data is obtained by classifying a history of past operation data in the refrigerator into a plurality of groups by clustering and assigning data indicating whether or not the heat exchanger is frosted to each group, and information indicating whether or not the heat exchanger is frosted predicted by the learning completion model is output.

Description

Technical Field

[0001] The present disclosure relates to a prediction device, a refrigeration system, a prediction method and a prediction program. Background Art

[0002] As an example of a defrosting process in a refrigeration cycle device, for example, Patent Document 1 listed below discloses determining a start timing of a defrosting operation using reinforcement learning.

[0003] <Prior Art Literature>

[0004] <Patent Documents>

[0005] Patent Document 1: International Publication No. 2021 / 176689 Summary of the invention

[0006] <Problems to be Solved by the Invention>

[0007] However, in the case of the defrosting process using reinforcement learning as described above, learning for improving the accuracy of determination of the start timing takes time and requires huge amounts of data.

[0008] An object of the present disclosure is to enable the defrosting operation to be started at a timing corresponding to frost formation.

[0009] <Methods used to solve the problem>

[0010] The first aspect of the present disclosure provides a prediction device that outputs information for controlling the start timing of a defrosting operation on a heat exchanger that functions as a heat absorber through a refrigerator, the refrigerator having a refrigerant circuit that connects a compressor, a radiator, an expansion valve, and the heat absorber in a ring shape and circulates refrigerant, and a control unit of the prediction device, when generating a learning completion model based on learning data, inputs current operating data of the refrigerator into the learning completion model, the learning data being obtained by clustering historical records of past operating data in the refrigerator into a plurality of groups and assigning data indicating whether or not frosting is formed on the heat exchanger to each group, and outputs information indicating whether or not frosting is formed on the heat exchanger predicted by the learning completion model.

[0011] According to the first aspect of the present disclosure, the defrosting operation can be started at a timing corresponding to frost formation.

[0012] Furthermore, a second aspect of the present disclosure is the prediction device according to the first aspect, wherein the operation data includes data indicating a state of the refrigerant circuit.

[0013] In addition, a third aspect of the present disclosure is the prediction device according to the second aspect, wherein the operation data includes data indicating a state of an environment.

[0014] In addition, the fourth aspect of the present disclosure is a prediction device according to the third aspect, wherein the data representing the state of the refrigerant circuit includes any one of the frequency of the compressor, the pressure data of the low-pressure side of the compressor, the opening data of the expansion valve, the heating capacity data, and the evaporation temperature data, and the data representing the state of the environment includes environmental humidity data.

[0015] In addition, aspect 5 of the present disclosure is a prediction device according to any one of aspects 1 to 4, wherein the control unit outputs information indicating the start timing of a defrosting action as information indicating whether or not frost is formed on the heat exchanger when frost is predicted by the learning completion model.

[0016] In addition, aspect 6 of the present disclosure is a prediction device according to any one of aspects 1 to 5, wherein the control unit generates the learning data by clustering the historical records of past operating data in the refrigerator into a plurality of groups and assigning data indicating whether or not frost is formed on the heat exchanger to each group.

[0017] In addition, the 7th aspect of the present disclosure is a prediction device according to any one of aspects 1 to 6, wherein the control unit learns the learning model in such a manner that the historical records of the past operating data classified into groups by the learning data are input into the learning model so that the output data output from the learning model is close to the data indicating whether or not frost is formed on the heat exchanger assigned to the classified group, thereby generating the learning completion model.

[0018] Aspect 8 of the present disclosure provides a refrigeration system having the above-mentioned refrigerator, wherein the refrigerator uses the information indicating whether frost is formed on the heat exchanger output by the prediction device according to any one of aspects 1 to 7 to control the start timing of the defrosting action.

[0019] Moreover, a ninth aspect of the present disclosure is the refrigeration system according to the eighth aspect, wherein the prediction device is implemented in any one of a plurality of units included in the refrigerating machine.

[0020] In addition, a tenth aspect of the present disclosure is a refrigeration system according to the eighth aspect, comprising a server device connected to the refrigeration machine via a network, wherein the prediction device is implemented in the server device.

[0021] The 11th aspect of the present disclosure provides a prediction method of a prediction device, the prediction device outputs information for controlling the start timing of a defrosting action of a heat exchanger functioning as a heat absorber by a refrigerator, the refrigerator having a refrigerant circuit, the refrigerant circuit connecting a compressor, a radiator, an expansion valve and the heat absorber in a ring shape, and circulating the refrigerant, the prediction method having the following steps: when a learning completion model is generated based on learning data, a step of inputting current operating data in the refrigerator into the learning completion model, the learning data being obtained by clustering historical records of past operating data in the refrigerator into a plurality of groups and assigning data indicating whether the heat exchanger is frosted to each group; and a step of outputting information indicating whether the heat exchanger is frosted as predicted by the learning completion model.

[0022] A twelfth aspect of the present disclosure provides a prediction program of a prediction device, the prediction device outputting information for controlling the start timing of a defrosting action on a heat exchanger functioning as a heat absorber by a refrigerator, the refrigerator having a refrigerant circuit, the refrigerant circuit connecting a compressor, a radiator, an expansion valve and the heat absorber in a ring shape and circulating the refrigerant, the prediction program being used to cause a control unit of the prediction device to perform the following steps: when a learning completion model is generated based on learning data, a step of inputting current operating data in the refrigerator into the learning completion model, the learning data being obtained by clustering historical records of past operating data in the refrigerator into a plurality of groups and assigning data indicating whether or not frosting is formed on the heat exchanger to each group; and a step of outputting information indicating whether or not frosting is formed on the heat exchanger predicted by the learning completion model. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] [ Figure 1A ] Figure 1A 1 is a diagram showing an example of the system configuration of the refrigeration system in the learning phase.

[0024] [ Figure 1B ] Figure 1B This is a diagram showing an example of a system configuration of a system related to an operation data collection process in a refrigeration system in a learning phase.

[0025] [ Figure 2 ] Figure 2 is a diagram showing an example of the hardware configuration of a learning device.

[0026] [ Figure 3 ] Figure 3 This is a diagram showing an example of a functional configuration of a system related to a learning process in a refrigeration system in a learning phase.

[0027] [ Figure 4 ] Figure 4 It is a diagram showing an example of operating data and observation data.

[0028] [ Figure 5 ] Figure 5 It is a diagram showing a specific example of the processing of the cluster analysis unit.

[0029] [ Figure 6 ] Figure 6 2 is a diagram showing a specific example of the processing of the labeling unit.

[0030] [ Figure 7 ] Figure 7 This is a diagram showing a specific example of the processing of the learning unit.

[0031] [ Figure 8 ] Figure 8 : is a flowchart showing the flow of the learning process.

[0032] [ Fig. 9 ] Fig. 9 This is a diagram showing an example of the system configuration of a refrigeration system in the prediction stage.

[0033] [ Fig.10 ] Fig.10 This is a diagram showing an example of the hardware configuration of a server device.

[0034] [ Fig.11 ] Fig.11 This is a diagram showing an example of the functional configuration of a system related to the prediction process in the refrigeration system in the prediction stage.

[0035] [ Fig.12 ] Fig.12 : is a flowchart showing the flow of prediction processing.

[0036] [ Fig.13 ] Fig.13 2 is a diagram showing a specific example of the defrosting process. DETAILED DESCRIPTION

[0037] Hereinafter, each embodiment will be described with reference to the drawings. It should be noted that in this specification and the drawings, components having substantially the same functional configuration are denoted by the same reference numerals to omit repeated descriptions.

[0038] [First embodiment]

[0039] <System configuration of refrigeration system in the learning stage>

[0040] First, the system configuration of the refrigeration system according to the first embodiment in the learning phase will be described. Figure 1A 1 is a diagram showing an example of the system configuration of the refrigeration system in the learning phase.

[0041] like Figure 1A As shown, the refrigeration system 100 according to the first embodiment includes a refrigerating machine (a refrigerating machine composed of a plurality of units including an indoor unit 110, an indoor unit 120, and an outdoor unit 130) and a learning device 150 in the learning stage. It should be noted that in this embodiment, a case where two indoor units are provided as a refrigerating machine is described as an example, but the number of indoor units may be one or more than three.

[0042] The refrigerator includes a refrigerant circuit 140 in which a refrigerant circulates, and a compressor, a radiator, an expansion valve, and a heat absorber are connected in a ring shape along the refrigerant circuit 140 to achieve indoor air conditioning.

[0043] It should be noted that, during the learning phase, the refrigeration system 100 collects operational data related to frost formation in the outdoor unit 130 (specifically, the heat exchanger 144) during the operation of the refrigerator. Figure 1A In the illustrated system configuration, devices required for collecting operation data related to frost formation in the heat exchanger 144 of the outdoor unit 130 are described, and devices required for the operation of the refrigerator itself are described in a simplified manner.

[0044] The outdoor unit 130 includes an electronic expansion valve 141 , a plate heat exchanger 142 , a liquid receiver 143 , a heat exchanger 144 , a flow path switching mechanism 145 , a compression mechanism 146 , a pressure sensor 147 , an ambient humidity sensor 148 , and an outdoor controller 149 .

[0045] The electronic expansion valve 141 adjusts the amount of refrigerant supplied to the indoor units 110 and 120 between the heat exchangers 111 and 121 when they function as heat absorbers and the heat exchanger 144 when they function as radiators (i.e., during cooling or cooling operation). Alternatively, the electronic expansion valve 141 adjusts the amount of refrigerant passing between the heat exchangers 111 and 121 when they function as radiators and the heat exchanger 144 when they function as heat absorbers (i.e., during heating or warming operation). The plate heat exchanger 142 cools the refrigerant. The liquid receiver 143 stores the refrigerant.

[0046] The heat exchanger 144 is an air heat exchanger that performs heat exchange between outdoor air sent by an outdoor fan (not shown) and the refrigerant.

[0047] The temperature sensor 144t measures the evaporation temperature of the refrigerant in the heat exchanger 144 when the heat exchanger 144 functions as a heat absorber.

[0048] The flow path switching mechanism 145 switches the flow path of the refrigerant. For example,

[0049] During cooling or refrigeration operation, the flow switching mechanism 145 draws the refrigerant evaporated (absorbs heat) in the heat exchanger 111 of the indoor unit 110 and the heat exchanger 121 of the indoor unit 120 into the compression mechanism 146. In addition, the flow switching mechanism 145 transfers the refrigerant compressed in the compression mechanism 146 to the heat exchanger 144. Thus, the refrigerant dissipates heat in the heat exchanger 144, and after the supply amount is adjusted by the electronic expansion valve 141, it evaporates again in the heat exchanger 111 of the indoor unit 110 and the heat exchanger 121 of the indoor unit 120.

[0050] During heating or warming operation, the flow switching mechanism 145 transfers the refrigerant compressed in the compression mechanism 146 to the indoor units 110 and 120. As a result, the refrigerant dissipates heat in the heat exchangers 111 and 121, and after the flow rate is adjusted by the electronic expansion valve 141, it evaporates (absorbs heat) in the heat exchanger 144. In addition, the flow switching mechanism 145 draws the refrigerant evaporated (absorbs heat) in the heat exchanger 144 into the compression mechanism 146. As a result, the refrigerant is compressed again in the compression mechanism 146.

[0051] During the defrosting operation, the flow switching mechanism 145 operates in the same manner as in the cooling or refrigeration operation. That is, the refrigerant compressed in the compression mechanism 146 dissipates heat in the heat exchanger 144. As a result, the frost on the surface of the heat exchanger 144 is heated from the inside, and the heat exchanger 144 is defrosted.

[0052] The compression mechanism 146 includes a plurality of compressors ( Figure 1A In the example of FIG. 1 , three compressors (compressors 146_1 to 146_3) compress the refrigerant. The plurality of compressors are variable capacity compressors whose frequency or rotation speed can be adjusted during operation.

[0053] The pressure sensor 147 measures the pressure of the refrigerant (low-pressure side pressure) at the inlet of the compression mechanism 146. The environmental humidity sensor 148 measures the humidity of the external environment.

[0054] The outdoor controller 149 controls the operation of each device included in the outdoor unit 130. In addition, the outdoor controller 149 collects measurement data of each sensor included in the outdoor unit 130 and the like.

[0055] The indoor unit 110 is a utilization unit for performing indoor air conditioning, and includes a heat exchanger 111 , a temperature sensor 112 , and an indoor controller 113 .

[0056] The heat exchanger 111 is an air heat exchanger that performs heat exchange between indoor air sent by an indoor fan (not shown) and a refrigerant.

[0057] The temperature sensor 122 measures the evaporation temperature of the refrigerant in the heat exchanger 111 when the heat exchanger 111 functions as a heat absorber.

[0058] The indoor controller 113 controls the operation of each device included in the indoor unit 110. In addition, the indoor controller 113 obtains measurement data of each sensor included in the indoor unit 110 and notifies the outdoor controller 149 of the measurement data.

[0059] The indoor unit 120 is a utilization unit that cools indoor air, and includes a heat exchanger 121 , a temperature sensor 122 , and an indoor controller 123 .

[0060] The heat exchanger 121 is an air heat exchanger that performs heat exchange between indoor air sent by an indoor fan (not shown) and the refrigerant.

[0061] The temperature sensor 112 measures the evaporation temperature of the refrigerant in the heat exchanger 121 when the refrigerant functions as a heat absorber.

[0062] The indoor controller 123 controls the operation of each device included in the indoor unit 120. In addition, the indoor controller 123 obtains measurement data of each sensor included in the indoor unit 120 and notifies the outdoor controller 149 of the measurement data.

[0063] The learning device 150 acquires operation data related to frost formation in the heat exchanger 144 of the outdoor unit 130 collected by the outdoor controller 149 during the operation of the refrigerator. In addition, the learning device 150 acquires observation data indicating that the heat exchanger 144 of the outdoor unit 130 is actually frosted (or not frosted) at each time when the acquired operation data is measured. In addition, the learning device 150 generates a learning completion model using the operation data and the observation data.

[0064] <Regarding the System Configuration of the System Related to Operation Data Collection and Processing>

[0065] Next, a system configuration of a system related to an operation data collection process for collecting operation data related to frost formation in the heat exchanger 144 of the outdoor unit 130 in the refrigeration system 100 during the learning phase will be described. Figure 1B This is a diagram showing an example of a system configuration of a system related to an operation data collection process in a refrigeration system in a learning phase.

[0066] like Figure 1BAs shown, the system related to the operation data collection and processing includes an indoor controller 113, an outdoor controller 149, and an indoor controller 123.

[0067] The indoor controller 113 includes a microcomputer 161 mounted on a control substrate, and a memory device 162 storing a program for operating the microcomputer 161 .

[0068] Similarly, the indoor controller 123 includes a microcomputer 181 mounted on a control substrate, and a memory device 182 storing a program for operating the microcomputer 181 .

[0069] The outdoor controller 149 includes a microcomputer 171 mounted on a control substrate, a memory device 172 storing a program for operating the microcomputer 171, and a communication device 173. In the outdoor controller 149, during the operation of the refrigerator, the outdoor controller 149 collects

[0070] The valve opening of the electronic expansion valve 141 (opening data of the expansion valve),

[0071] A total frequency (compressor frequency data) calculated based on the total rotation speed of the compressors 146_1 to 146_3,

[0072] The pressure of the refrigerant at the inlet of the compression mechanism 146 measured by the pressure sensor 147 (low-pressure side pressure data), the humidity of the external environment measured by the environmental humidity sensor 148 (environmental humidity data),

[0073] The heating capacity of the refrigerator (heating capacity data),

[0074] The evaporation temperature of the refrigerant in the heat exchanger 144 measured by the temperature sensor 144t (evaporation temperature data).

[0075] Thus, in the outdoor controller 149 according to the present embodiment, as the operation data related to frost formation, the low-pressure side pressure data, which are data indicating the state of the refrigerant circuit 140,

[0076] ·Expansion valve opening data,

[0077] Frequency data of the compressor,

[0078] Heating capacity data,

[0079] Evaporation temperature data,

[0080] and data representing the state of the environment

[0081] Environmental humidity data

[0082] The collected operation data can be sent to the learning device 150 via the communication device 173. In addition, in the present embodiment, the microcomputer and the memory device included in each controller of the refrigerator may be collectively referred to as a "control unit".

[0083] <Hardware Configuration of Learning Device>

[0084] Next, the hardware configuration of the learning device 150 will be described. Figure 2 is a diagram showing an example of the hardware configuration of a learning device.

[0085] like Figure 2 As shown, the learning device 150 includes a processor 201, a memory 202, an auxiliary storage device 203, a display device 204, an operation device 205, and a communication device 206. It should be noted that the hardware components of the learning device 150 are connected to each other via a bus 207.

[0086] The processor 201 includes various computing devices such as a CPU (Central Processing Unit) and a GPU (Graphics Processing Unit). The processor 201 reads various programs (for example, a learning program described later) to the memory 202 and executes them.

[0087] The memory 202 includes a main storage device such as a ROM (Read Only Memory) or a RAM (Random Access Memory). The processor 201 and the memory 202 form a so-called computer, and the processor 201 executes various programs read from the memory 202 to cause the computer to realize various functions.

[0088] The auxiliary storage device 203 stores various programs or various data used when the processor 201 executes the various programs.

[0089] The display device 204 is a display device that displays the internal processing results of the learning device 150. The operation device 205 is an input device used when the user inputs various instructions to the learning device 150.

[0090] The communication device 206 is connected to, for example, the outdoor controller 149 of the refrigerator, and receives operation data.

[0091] <Regarding the System Configuration of the System Related to the Learning Process>

[0092] Next, the functional configuration of the system related to the learning process in the refrigeration system 100 in the learning phase will be described. Figure 3This is a diagram showing an example of a functional configuration of a system related to a learning process in a refrigeration system in a learning phase.

[0093] As described above, the outdoor controller 149 functions as the operation data collection unit 301 in the learning phase. The operation data collection unit 301 collects operation data during the operation of the refrigerator and transmits the operation data to the learning device 150 .

[0094] As described above, the learning device 150 has a learning program installed therein, and by executing the program, the learning device 150 functions as the data acquisition unit 311 , the cluster analysis unit 312 , the labeling unit 313 , and the learning unit 314 .

[0095] The data acquisition unit 311 acquires the operation data transmitted from the outdoor controller 149. Specifically, the data acquisition unit 311 acquires low-pressure side pressure data, which is data indicating the state of the refrigerant circuit 140 at each time.

[0096] ·Expansion valve opening data,

[0097] Frequency data of the compressor,

[0098] Heating capacity data,

[0099] Evaporation temperature data,

[0100] and data representing the state of the environment at each time

[0101] Environmental humidity data

[0102] As operating data.

[0103] In addition, the data acquisition unit 311 acquires observation data. As described above, the observation data is data indicating whether or not frost is formed on the heat exchanger 144 at each time when the operation data is measured, and the result of observation of whether or not frost is formed by an observer is input as the observation data.

[0104] It should be noted that the data acquisition unit 311 stores the acquired operating data and observation data in the data storage unit 321 .

[0105] The cluster analysis unit 312 reads the operation data from the data storage unit 321 and performs cluster analysis based on, for example, the k-means method, thereby classifying the operation data at each time into a plurality of groups. As described above, the operation data includes six types of data (low-pressure side pressure data, expansion valve opening data, compressor frequency data, heating capacity data, evaporation temperature data, and ambient humidity data). Therefore, the cluster analysis unit 312 groups the operation data at each time scattered in the 6-dimensional feature space in such a way that operation data with close distribution positions belong to the same group.

[0106] In addition, the cluster analysis unit 312 notifies the labeling unit 313 of the operation data at each time that is classified into each group.

[0107] The labeling unit 313 adds a label to the operation data at each time classified into each group. Specifically, the labeling unit 313 reads the observation data from the data storage unit 321 and determines whether the operation data at each time classified into each group is the operation data at the time when frost is observed or the operation data at the time when no frost is observed.

[0108] Furthermore, when the operation data at each time of the group classified as the processing target is the operation data at a time when frost is observed, the labeling unit 313 adds a frost label to all the operation data at each time of the group classified as the processing target.

[0109] Furthermore, when the operation data at each time of the group classified as the processing target is the operation data at a time when no frost is observed, the labeling unit 313 adds a no frost label to all the operation data at each time of the group classified as the processing target.

[0110] Furthermore, the labeling unit 313 stores the operation data at each time to which the label of frost formation or the label of no frost formation is added as learning data in the learning data storage unit 322 .

[0111] The learning unit 314 performs learning on the learning model using the learning data stored in the learning data storage unit 322 , and generates a learned model.

[0112] <Specific Examples of Operation Data and Observation Data>

[0113] Next, specific examples of the operating data and observation data acquired by the data acquisition unit 311 will be described. Figure 4 It is a diagram showing an example of operating data and observation data.

[0114] like Figure 4 As shown, the operation data 410 includes data items (low-pressure side pressure, expansion valve opening, compressor frequency, heating capacity, evaporation temperature, ambient humidity) as information items. In addition, the operation data 410 includes time (t1, t2, t3, t4, t5, ..., t x ) as an item of information.

[0115] In each column of the operation data 410, operation data at each time is stored. For example, in the column where data item = "low-pressure side pressure" and time = "t1" are interlaced, low-pressure side pressure data at time t1 is stored. Similarly, in the column where data item = "opening of expansion valve" and time = "t1" are interlaced, expansion valve opening data at time t1 is stored. Similarly, in the column where data item = "frequency of compressor" and time = "t1" are interlaced, compressor frequency data at time t1 is stored. Similarly, in the column where data item = "heating capacity" and time = "t1" are interlaced, heating capacity data at time t1 is stored. Similarly, in the column where data item = "evaporation temperature" and time = "t1" are interlaced, evaporation temperature data at time t1 is stored. Similarly, in the column where data item = "ambient humidity" and time = "t1" are interlaced, ambient humidity data at time t1 is stored.

[0116] It should be noted that if Figure 4 As shown, the operation data at time "t1" is plotted as point P1 in the six-dimensional feature quantity space.

[0117] Hereinafter, similarly, at each time (t2, t3, t4, t5, ..., t x ) are plotted as point P2, point P3, point P4, point P5, ..., point P x .

[0118] In addition, if Figure 4 As shown, the observation data 420 includes a data item (presence or absence of frost) as an information item. In addition, the observation data 420 includes time (t1, t2, t3, t4, t5, ..., t x ) as an item of information.

[0119] In each column of the observation data 420, observation data at each time is stored. For example, in a column where data item = "presence of frost" and time = "t1" are interlaced, data indicating whether frost is present at time t1 is stored. Figure 4 In the example of FIG. 1 , the column indicated by hatching is data indicating the presence of frost, and the column not indicated by hatching is data indicating the absence of frost.

[0120] <Specific Example of Processing by Cluster Analysis Unit>

[0121] Next, a specific example of the processing of the cluster analysis unit 312 will be described. Figure 5 It is a diagram showing a specific example of the processing of the cluster analysis unit.

[0122] As described above, the cluster analysis unit 312 reads the operation data at each time from the data storage unit 321 and classifies the operation data into a plurality of groups by performing cluster analysis.

[0123] exist Figure 5 In the feature quantity space 510, the operation data of each time period read out from the data storage unit 321 (point P1, point P2, point P3, point P4, point P5, ..., P x It should be noted that, for the sake of convenience, the feature quantity space 510 is represented as a three-dimensional feature quantity space (low-pressure side pressure data, compressor frequency data, heating capacity data).

[0124] The cluster analysis unit 312 groups the points distributed in the feature value space 510 so that the operating data with close distribution positions belong to the same group.

[0125] exist Figure 5 In FIG. 5 , the feature space 520 shows a situation where group 1 and group 2 are generated by grouping. It should be noted that the cluster analysis unit 312 classifies all points distributed in the feature space 510 into any one group.

[0126] <Specific Example of Processing of Marking Portion>

[0127] Next, a specific example of the processing of the labeling unit 313 will be described. Figure 6 2 is a diagram showing a specific example of the processing of the labeling unit.

[0128] As described above, the labeling unit 313 adds labels to the operation data of each time that is classified into each group. Figure 6 In FIG. 6 , the clustering result 610 is a list of points indicating the operation data at each time that are classified into each group. Figure 6 The example of shows a case where the operation data at each time is classified into α groups of group 1 to group α.

[0129] In addition, Figure 6 In FIG. 6 , the labeling result 620 shows that a label indicating that frost is present or a label indicating that frost is not present is added to the operation data at each time period classified into each group. Figure 6 The example shows a case where a label "1" indicating frost formation is added to the operation data at each time period classified into group 1, group 3, and group 4, and a label "0" indicating no frost formation is added to the operation data at each time period classified into group 2 and group α.

[0130] In this way, by labeling the operating data in groups generated by cluster analysis, the same effect as when the learning data is generated even for the operating data at positions not plotted in the 6-dimensional feature space can be obtained. Therefore, in the learning unit 314, a learning model with high prediction accuracy can be generated with less learning data and less learning time.

[0131] <Specific example of the processing of the learning unit>

[0132] Next, a specific example of the processing of the learning unit 314 will be described. Figure 7 This is a diagram showing a specific example of the processing of the learning unit.

[0133] As described above, the learning unit 314 uses the learning data stored in the learning data storage unit 322 to learn the learning model and generate a learning completed model. Figure 7 In FIG. 1 , learning data 700 shows an example of learning data read from the learning data storage unit 322. Figure 7 As shown, in the learning data 700 , labels are given to the operating data at each time.

[0134] The learning unit 314 inputs the operation data at each time in the learning data 700 as input data to the random forest as an example of a learning model. In addition, the learning unit 314 inputs the label at each time in the learning data 700 as true value data to the random forest as an example of a learning model. Thus, in the learning unit 314, learning can be performed on the random forest in such a way that the output data output by the random forest when the input data is inputted is close to the true value data.

[0135] Therefore, the learning unit 314 can generate a learned model that has learned the relationship between the operating data and the presence or absence of frost on the heat exchanger 144 .

[0136] <Flow of learning process>

[0137] Next, the flow of the learning process performed by the learning device 150 will be described. Figure 8 : is a flowchart showing the flow of the learning process.

[0138] In step S801 , the learning device 150 acquires operating data and observation data.

[0139] In step S802 , the learning device 150 performs cluster analysis on the acquired operating data, and classifies the operating data at each time into a plurality of groups.

[0140] In step S803 , the learning device 150 adds a frosting label or a no-frosting label to the operation data of each time period classified into each group based on the acquired observation data.

[0141] In step S804 , the learning device 150 performs learning on the learning model using the operation data at each time to which the labels are added, and generates a learned model.

[0142] <System configuration of refrigeration system in the prediction stage>

[0143] Next, the system configuration in the prediction phase of the refrigeration system according to the first embodiment will be described. Fig. 9 This is a diagram showing an example of the system configuration of a refrigeration system in the prediction stage.

[0144] and use Figure 1A The refrigeration system 100 in the described learning phase is different from the refrigeration system 900 in the prediction phase in that the refrigeration machine is connected to the server device 910 via the network 920 .

[0145] The server device 910 is an example of a prediction device, and receives operation data from the outdoor unit 130 of the operating refrigerator via the network 920. In addition, based on the received operation data, the server device 910 predicts whether or not frost is formed on the heat exchanger 144 of the outdoor unit 130, and transmits information indicating the presence of frost to the outdoor unit 130 if frost is predicted.

[0146] In this way, by configuring so as to predict the presence or absence of frost in the heat exchanger 144 based on the operation data, the defrosting operation can be started in the refrigerator based on the presence or absence of frost.

[0147] It should be noted that the defrosting operation refers to the operation of each device of the refrigerator for performing the above-mentioned defrosting operation.

[0148] <Hardware Configuration of Server Device>

[0149] Next, the hardware configuration of the server device 910 will be described. Fig.10 This is a diagram showing an example of the hardware configuration of a server device.

[0150] like Fig.10 As shown, the server device 910 includes a processor 1001 , a memory 1002 , an auxiliary storage device 1003 , a display device 1004 , an operating device 1005 , and a communication device 1006 . It should be noted that the hardware components of the server device 910 are connected to each other via a bus 1007 .

[0151] The processor 1001 includes various computing devices such as a CPU (Central Processing Unit) and a GPU (Graphics Processing Unit). The processor 1001 reads various programs (for example, a prediction program described later) to the memory 1002 and executes them.

[0152] The memory 1002 has a main storage device such as a ROM (Read Only Memory) and a RAM (Random Access Memory). The processor 1001 and the memory 1002 form a so-called computer (sometimes referred to as a "control unit" in this embodiment), and the processor 1001 executes various programs read from the memory 1002, thereby enabling the computer to implement various functions.

[0153] The auxiliary storage device 1003 stores various programs and various data used when the processor 1001 executes the various programs.

[0154] The display device 1004 is a display device that displays the results of internal processing of the server device 910. The operation device 1005 is an input device used when the user inputs various instructions to the server device 910.

[0155] The communication device 1006 communicates with the outdoor controller 149 of the refrigerator via the network 920, for example.

[0156] <Functional Configuration of System Related to Prediction Processing>

[0157] Next, the functional configuration of the system related to the prediction process in the refrigeration system 900 in the prediction stage will be described. Fig.11 This is a diagram showing an example of the functional configuration of a system related to the prediction process in the refrigeration system in the prediction stage.

[0158] like Fig.11 As shown in (a), the outdoor controller 149 functions as an operation data collection unit 1101 and a defrost control unit 1102 during the operation of the refrigerator in the prediction phase.

[0159] As described above, the prediction program is installed in the server device 910, and by executing the program, the server device 910 functions as the data acquisition unit 1111 and the learned model 1112 during the operation of the refrigerator.

[0160] The operation data collection unit 1101 collects the operation data and transmits the collected operation data to the server device 910 via the network 920 .

[0161] The data acquisition unit 1111 acquires operation data from the outdoor controller 149 of the refrigeration system 900. Specifically, the data acquisition unit 1111 acquires low-pressure side pressure data, which is data indicating the state of the refrigerant circuit 140 at each time.

[0162] ·Expansion valve opening data,

[0163] Frequency data of the compressor,

[0164] Heating capacity data,

[0165] Evaporation temperature data,

[0166] and data representing the state of the environment at each time

[0167] Environmental humidity data

[0168] As operating data.

[0169] The learned model 1112 predicts the presence or absence of frost on the heat exchanger 144 at each time by inputting the operation data at each time acquired by the data acquisition unit 1111. When frost is predicted, the learned model 1112 sends information indicating the presence of frost to the outdoor controller 149.

[0170] When receiving the information indicating the presence of frost from the server device 910, the defrost control unit 1102 controls the heat exchanger 144 functioning as a heat absorber to start the defrosting operation. That is, in this embodiment, the information indicating the presence of frost sent from the server device 910 is information indicating the start timing of the defrosting operation of the heat exchanger 144 functioning as a heat absorber.

[0171] It should be noted that Fig.11 (a) shows a functional configuration in which the refrigeration system 900 includes a server device 910 and a prediction program is installed in the server device 910 so that the control unit of the server device 910 predicts whether or not frost is formed in the heat exchanger 144 .

[0172] However, in the refrigeration system 900, the installation destination of the prediction program is not limited to the server device 910. For example, an information processing device may be provided near the refrigeration machine, and the prediction program may be installed in the information processing device. In this case, the information processing device functions as a prediction device and has the same Fig.11 The server device 910 of (a) has the same functional configuration.

[0173] The prediction program may be installed in, for example, the outdoor controller 149, the indoor controller 113, or the indoor controller 123 of the refrigerator. In this case, the control unit of the controller installed in the prediction program functions as the prediction device.

[0174] Fig.11 (b) shows the functional configuration of the outdoor controller 149 when the prediction program is installed in the outdoor controller 149. Fig.11 As shown in (b), by executing the prediction program in the outdoor controller 149, the control unit of the outdoor controller 149 functions as an operation data collection unit 1101, a learning completion model 1112, and a defrost control unit 1102. It should be noted that the details of the functions of each unit have been explained, so the explanation is omitted here.

[0175] <Flow of prediction processing>

[0176] Next, the flow of prediction processing performed by the server device 910 will be described. Fig.12 : is a flowchart showing the flow of prediction processing.

[0177] In step S1201, the server device 910 acquires the operating data. In step S1202, the server device 910 uses the learned model to predict the presence or absence of frost on the heat exchanger 144 based on the acquired operating data.

[0178] In step S1203, server device 910 determines whether frost is predicted by the learning model. If frost is predicted (if "yes" in step S1203), the process proceeds to step S1204. In step S1204, server device 910 transmits information indicating frost to the refrigerator.

[0179] On the other hand, if it is determined that there is no frost (if "No" in step S1203), the process proceeds to step S1205. In step S1205, the server device 910 determines whether to end the prediction process. In step S1205, if it is determined that the prediction process is to be continued (if "No" in step S1205), the process returns to step S1201. On the other hand, if it is determined that the prediction process is to be ended (if "Yes" in step S1205), the prediction process is ended.

[0180] <Specific example of defrosting process>

[0181] Next, a specific example of the defrosting process performed by the refrigeration system 900 will be described. Fig.13 2 is a diagram showing a specific example of the defrosting process.

[0182] exist Fig.13 In the graph 1300, the horizontal axis represents time, the vertical axis on the left represents the start timing of the defrosting action for the heat exchanger 144 functioning as a heat absorber, and the vertical axis on the right represents the operation data. In addition, each line of the graph represents the low-pressure side pressure data, the frequency data of the compressor, and the heating capacity data. In addition, the thick line of the graph represents the start timing of the defrosting action for the heat exchanger 144 functioning as a heat absorber.

[0183] like Fig.13 As shown, in the case of the refrigeration system 900 according to the present embodiment, the learning model that has learned the relationship between the operating data and the presence or absence of frost on the heat exchanger 144 is used to predict the presence or absence of frost on the heat exchanger 144 based on the operating data. In addition, in the case of the refrigeration system 900 according to the present embodiment, when it is predicted that there is frost, the defrosting action is started. Therefore, according to the refrigeration system 900 according to the present embodiment, unlike the case where the defrosting action is started based on the operating data regardless of whether there is actual frost, the defrosting action can be started only when there is actual frost. Therefore, according to the refrigeration system 900 according to the present embodiment, the number of defrosting actions can be reduced.

[0184] <Conclusion>

[0185] As can be seen from the above description, in the refrigeration system according to the first embodiment,

[0186] A refrigerator is provided, the refrigerator having a refrigerant circuit in which a compressor, a radiator, an expansion valve, and the heat absorber are connected in an annular shape and a refrigerant circulates.

[0187] In the learning phase, a learned model is generated based on learning data obtained by classifying the history of past operation data in the refrigerator into a plurality of groups by clustering and assigning data indicating the presence or absence of frost on the heat exchanger to each group.

[0188] In the prediction stage, the current operating data of the refrigerator is input to the learned model, and information indicating the presence or absence of frost on the heat exchanger predicted by the learned model is output.

[0189] Thus, in the first embodiment, the learning model that has learned the relationship between the operation data and the presence or absence of frost on the heat exchanger is used to predict and output the presence or absence of frost on the heat exchanger based on the operation data. Thus, according to the first embodiment, the defrosting operation can be started for the heat exchanger that functions as a heat absorber based on the information indicating the presence of frost. Therefore, according to the first embodiment, the refrigerator can start the defrosting operation at a timing corresponding to frost.

[0190] [Second embodiment]

[0191] In the first embodiment described above, a case where one learning completed model 1112 is generated using the learning data 700 is described. However, the number of learning completed models generated using the learning data 700 is not limited to one. For example, the learning data 700 may be divided into each season, and learning completed models may be generated using the learning data corresponding to each season.

[0192] Alternatively, the learning data 700 may be divided for each region (climate) where a refrigerator is installed, and the learned models may be generated using the learning data corresponding to each region (climate).

[0193] In addition, in the first embodiment described above, the server device 910 is configured to send information indicating the presence of frost to the refrigerator when it is determined that frost is present, and in the refrigerator, at the timing when the information indicating the presence of frost is received, the heat exchanger functioning as a heat absorber starts to perform a defrosting operation. However, it is not necessary to use the information indicating the presence of frost as information indicating the start timing of defrosting, and the refrigerator may be configured to start the defrosting operation, for example, based on the satisfaction of other conditions. That is, the information indicating the presence of frost output by the server device 910 can be used not only as information indicating the start timing of the defrosting operation, but also as "information for controlling the start timing of the heat exchanger functioning as a heat absorber when the defrosting operation starts".

[0194] In the first embodiment, six types of data are exemplified as the operation data, but the operation data is not limited to these six types of data, as long as at least one of the six types of data is included. In addition, the operation data may include data other than the six types of data. For example, the data representing the state of the environment in the operation data may include environmental temperature data in addition to environmental humidity data.

[0195] In addition, in the first embodiment, the case of predicting whether or not frost is formed on the heat exchanger 144 (more precisely, the heat transfer fins) of the outdoor unit 130 is described. However, the object of predicting whether or not frost is formed is not limited to the heat exchanger 144 of the outdoor unit 130, and for example, whether or not frost is formed on the indoor unit 120 may be predicted. In this case, the refrigeration system 900 may be configured to use the same operating data as the case of predicting whether or not frost is formed on the outdoor unit 130, or may be configured to use operating data different from that in this case.

[0196] In the first embodiment described above, the case where the learning device and the prediction device are configured as independent devices is described. However, the learning device and the prediction device may be configured as an integrated device. For example, the server device 910, which is an example of a prediction device, may be configured to perform both the learning process and the prediction process. In this case, the control unit of the server device 910 performs the learning process (see Figure 3 ) and prediction processing (refer to Fig.11 ) functions. In addition, for example, in a refrigerator as an example of a prediction device, it is also possible to configure so that both the learning process and the prediction process are executed. In this case, the control unit of the refrigerator executes the learning process (see Figure 3 ) and prediction processing (refer to Fig.11 )The functions of both.

[0197] In addition, in the first embodiment described above, the learning stage and the prediction stage are shown, but the refrigeration system may be configured to further perform a re-learning stage. For example, the refrigeration system may be configured to re-learn the learning completed model by using the operating data collected in the prediction stage as learning data. It should be noted that, in the case of re-learning, the learning device and the prediction device may be provided as an integrated device, or may be provided as independent devices in the refrigeration system.

[0198] In addition, in the above-mentioned first embodiment, a case where a random forest is used as an example of a learning model is shown, but the learning model is not limited to a random forest, and any other learning model that performs supervised learning may be used.

[0199] Although the embodiments have been described above, it will be understood that various changes in form and detail may be made without departing from the spirit and scope of the claims.

[0200] This application claims priority based on Japanese patent application No. 2022-082891 filed on May 20, 2022, and all contents of the Japanese patent application are cited in this application by reference.

[0201] Explanation of symbols

[0202] 100: Refrigeration system

[0203] 110: Indoor unit

[0204] 111: Heat exchanger

[0205] 112: Temperature sensor

[0206] 113: Indoor controller

[0207] 120: Indoor unit

[0208] 121: Heat exchanger

[0209] 122: Temperature sensor

[0210] 123: Indoor controller

[0211] 130: Outdoor unit

[0212] 140: Refrigerant circuit

[0213] 141: Electronic expansion valve

[0214] 144: Heat exchanger

[0215] 144t: Temperature sensor

[0216] 146: Compression mechanism

[0217] 147: Pressure sensor

[0218] 148: Ambient humidity sensor

[0219] 149: Outdoor controller

[0220] 311: Data Acquisition Department

[0221] 312: Cluster Analysis Department

[0222] 313: Annotation Department

[0223] 314: Learning Department

[0224] 410: Operational data

[0225] 420: Observation Data

[0226] 510, 520: Feature space

[0227] 610: Clustering results

[0228] 620: Annotation results

[0229] 700: Learning Data

[0230] 900: Refrigeration system

[0231] 910: Server device

[0232] 1101: Operation data collection department

[0233] 1102: Defrost control unit

[0234] 1111: Data acquisition department

[0235] 1112: Learning completion model.

Claims

1. A prediction device that outputs information for controlling the start timing of a defrosting operation of a heat exchanger that functions as a heat absorber by a refrigerator, wherein the refrigerator has a refrigerant circuit that connects a compressor, a radiator, an expansion valve, and the heat absorber in a ring shape and circulates a refrigerant, The control unit of the prediction device When a learning model is generated based on learning data, current operating data of the refrigerator is input into the learning model, wherein the learning data is obtained by clustering the history of past operating data of the refrigerator into a plurality of groups and assigning data indicating whether or not frost is formed on the heat exchanger to each group. Information indicating whether or not frost is formed on the heat exchanger, predicted by the learned model, is output.

2. The prediction device according to claim 1, wherein: The operation data includes data indicating a state of the refrigerant circuit.

3. The prediction device according to claim 2, wherein: The operating data includes data indicating a state of an environment.

4. The prediction device according to claim 3, wherein: The data indicating the state of the refrigerant circuit includes any one of the frequency of the compressor, the pressure data of the low-pressure side of the compressor, the opening data of the expansion valve, the heating capacity data, and the evaporation temperature data. The data representing the state of the environment includes environmental humidity data.

5. The prediction device according to claim 1, wherein: The control unit outputs information indicating a start timing of a defrosting operation as information indicating whether or not frost is formed on the heat exchanger, when the learned model predicts the presence of frost.

6. The prediction device according to claim 1, wherein: The control unit generates the learning data by classifying the history of past operation data of the refrigerator into a plurality of groups by clustering and assigning data indicating the presence or absence of frost on the heat exchanger to each group.

7. The prediction device according to claim 1, wherein: The control unit generates the learning completion model by learning the learning model in such a manner that the historical records of the past operating data classified into groups as the learning data are input into the learning model so that the output data output from the learning model is close to the data indicating whether or not frost is formed on the heat exchanger assigned to the classified groups.

8. A refrigeration system comprising the refrigerator, wherein the refrigerator controls the start timing of a defrosting operation using the information indicating the presence or absence of frost on the heat exchanger output by the prediction device according to any one of claims 1 to 7.

9. The refrigeration system according to claim 8, wherein: The prediction device is implemented in any one of the plurality of units included in the refrigerator.

10. The refrigeration system according to claim 8, wherein: A server device connected to the refrigerator via a network, The prediction device is implemented in the server device.

11. A prediction method of a prediction device, the prediction device outputting information for controlling the start timing of a defrosting operation of a heat exchanger functioning as a heat absorber by a refrigerator, the refrigerator having a refrigerant circuit, the refrigerant circuit connecting a compressor, a radiator, an expansion valve, and the heat absorber in an annular shape and circulating a refrigerant, the prediction method comprising the following steps: A step of inputting current operation data of the refrigerator into the learned model when a learned model is generated based on learning data, wherein the learning data is obtained by clustering a history of past operation data of the refrigerator into a plurality of groups and assigning data indicating whether or not frost is formed on the heat exchanger to each group; and and outputting information indicating the presence or absence of frost on the heat exchanger predicted by the learned model.

12. A prediction program of a prediction device, the prediction device outputting information for controlling the start timing of a defrosting operation of a heat exchanger functioning as a heat absorber by a refrigerator, the refrigerator having a refrigerant circuit that connects a compressor, a radiator, an expansion valve, and the heat absorber in an annular shape and circulates a refrigerant, the prediction program causing a control unit of the prediction device to execute the following steps: A step of inputting current operation data of the refrigerator into the learned model when a learned model is generated based on learning data, wherein the learning data is obtained by clustering a history of past operation data of the refrigerator into a plurality of groups and assigning data indicating whether or not frost is formed on the heat exchanger to each group; and and outputting information indicating the presence or absence of frost on the heat exchanger predicted by the learned model.

Images