Refrigeration device and fault diagnosis method and system thereof
By correcting the operating data of the refrigeration device to be diagnosed to make it closer to the data under the preset environment, the problem of low diagnostic accuracy under the user environment is solved, and higher diagnostic accuracy and efficiency are achieved.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- HISENSE(SHANDONG)REFRIGERATOR CO LTD
- Filing Date
- 2024-11-20
- Publication Date
- 2026-05-22
AI Technical Summary
In the existing technology, the fault diagnosis model based on the preset environment has a low diagnostic accuracy when applied to the user environment, mainly because the temperature and humidity of the user environment are very different from the preset environment, and the food stored by the user is different.
By acquiring the operating data of the refrigeration unit to be diagnosed, correcting its key data to be the same as the data under normal operation in the preset environment, and then calling the fault diagnosis model trained in the preset environment to perform diagnosis, including the correction processing of the compressor's start-stop point temperature and start-stop cycle.
This improves the diagnostic accuracy of the fault diagnosis model in the user environment, ensuring the accuracy and efficiency of the diagnostic results.
Smart Images
Figure CN122072200A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of fault diagnosis technology, and in particular to a refrigeration device and its fault diagnosis method and system. Background Technology
[0002] Refrigeration devices such as refrigerators and freezers are consumer products that keep food or other items at a constant low temperature.
[0003] In related technologies, refrigerators typically include a cabinet with several compartments for storing food, beverages, or other items that need to be kept at low temperatures. Currently, fault diagnosis models are generally trained using laboratory simulation data to diagnose refrigerator operational data in the user's environment.
[0004] However, the refrigerator environment in a user's home differs significantly from that in a laboratory setting. Temperature and humidity vary greatly, regardless of whether it's the north or south, or summer or winter. Furthermore, the food stored in a user's refrigerator cannot be simulated, as the amount and type of food stored vary from person to person. This results in low accuracy when a fault diagnosis model trained under a preset environment is applied to diagnose faults in a user's refrigerator. Summary of the Invention
[0005] The main objective of this application is to propose a refrigeration device and its fault diagnosis method and system. When it is determined that the operating condition of the refrigeration device to be diagnosed is different from that of the refrigeration device under a preset environment, the key data in the operating data of the refrigeration device to be diagnosed is corrected to be the same as the key data in the normal operating data of the refrigeration device under the preset environment, and then the fault diagnosis model trained under the preset environment is called for diagnosis, which can improve the diagnostic accuracy.
[0006] To achieve the above objectives, this application proposes a fault diagnosis method for a refrigeration device, the method comprising:
[0007] Acquire the first operating data of the refrigeration device to be diagnosed within the acquisition period;
[0008] At least one key data is extracted from the first operating data, and the key data is compared with the corresponding key data in the second operating data. The second operating data is the data of the refrigeration device operating normally under a preset environment. The first operating data and the second operating data have the same collection period.
[0009] If the key data is different from the key data corresponding to the second running data, then the first running data is corrected according to the second running data so that the key data in the first running data is the same as the key data corresponding to the second running data.
[0010] The fault diagnosis model is invoked to diagnose the faults in the corrected first operating data and obtain the diagnosis results. The fault diagnosis model is obtained by simulating various faults of the refrigeration device under the preset environment and collecting and training data.
[0011] The above technical solution has the following advantages or beneficial effects:
[0012] When the key data in the first operating data of the refrigeration unit to be diagnosed is different from the corresponding key data in the second operating data of the refrigeration unit operating normally under the preset environment, it indicates that the first operating data may itself be fault data, or the first operating data is data under normal operation, but the operating condition is different from the operating condition under the preset environment. In this case, it is necessary to correct the key data in the first operating data to be the same as the corresponding key data in the second operating data, so as to make the operating data of the refrigeration unit to be diagnosed closer to the operating data of the refrigeration unit under the preset environment to a certain extent. Then, the fault diagnosis model trained under the preset environment can be called for diagnosis, thereby improving the diagnostic accuracy of the model. That is, when it is determined that the operating condition of the refrigeration unit to be diagnosed is inconsistent with the operating condition of the refrigeration unit under the preset environment, the operating data of the two are corrected to be closer, and then the fault diagnosis model is called for fault diagnosis, thereby improving the diagnostic accuracy of the fault diagnosis model in the actual application of the refrigeration unit to be diagnosed.
[0013] In one embodiment of this application, after calling a fault diagnosis model to perform fault diagnosis on the corrected first operating data and obtaining the diagnosis result, the method further includes:
[0014] If the diagnostic result indicates that the refrigeration device to be diagnosed has a malfunction, then the corresponding fault information is output.
[0015] If the diagnostic result indicates that the refrigeration device to be diagnosed is operating normally, then the subsequent operating data of the refrigeration device to be diagnosed is corrected according to the first correction processing method, and then the fault diagnosis model is called to perform fault diagnosis on the subsequent corrected operating data of the refrigeration device to be diagnosed.
[0016] The above technical solution has the following advantages or beneficial effects:
[0017] If the first operating data itself is fault data, the key data within it has been corrected to match the corresponding key parameters in the second operating data. This means the operating data of the refrigeration unit to be diagnosed has been corrected to closely approximate the operating data under a preset environment. Therefore, the fault diagnosis model trained under the preset environment can directly output accurate fault information. If the first operating data represents data from the normal operating condition of the refrigeration unit to be diagnosed, meaning the diagnosis result indicates that the refrigeration unit is operating normally without faults, then considering the possibility of future faults, the subsequent operating data of the refrigeration unit to be diagnosed needs to be corrected using the first correction method before calling the fault diagnosis model for diagnosis. In other words, it is no longer necessary to compare key data in the subsequent operating data of the refrigeration unit to be diagnosed; instead, the subsequent operating data can be directly corrected using the first correction method before directly calling the fault diagnosis model for fault diagnosis, thus improving fault diagnosis efficiency.
[0018] In one embodiment of this application, after comparing whether the key data is the same as the corresponding key data in the second running data, the method further includes:
[0019] If the key data is the same as the key data in the second operating data, then the fault diagnosis model is invoked to perform fault diagnosis on the subsequent operating data of the refrigeration device to be diagnosed, and a diagnosis result is obtained.
[0020] The above technical solution has the following advantages or beneficial effects:
[0021] When the key data in the first operating data of the refrigeration device to be diagnosed is the same as the key data in the second operating data of the refrigeration device operating normally under the preset environment, it can be determined that the refrigeration device to be diagnosed is operating normally and its operating condition is highly similar to that of the refrigeration device under the preset environment. At this time, the fault diagnosis model trained under the preset environment can be directly called for diagnosis, and the subsequent operating data of the refrigeration device to be diagnosed can be accurately used for fault diagnosis.
[0022] In one embodiment of this application, the key data includes the compressor's operating parameters. Correspondingly, comparing whether the key data is the same as the key data in the second operating data includes:
[0023] Compare the compressor operating parameters in the first operating data with those in the second operating data to see if they are the same.
[0024] The above technical solution has the following advantages or beneficial effects:
[0025] Considering that the compressor's operating status can intuitively and accurately reflect the operating status of the refrigeration unit, by extracting the compressor's operating parameters from the first operating data and determining whether the compressor's operating parameters in the first operating data are the same as those in the second operating data, it can be determined whether the compressor's operating status of the refrigeration unit to be diagnosed is consistent with the compressor's operating status under the preset environment. Thus, it can be determined more accurately whether the operating condition of the refrigeration unit to be diagnosed is basically the same as the operating condition of the refrigeration unit under the preset environment.
[0026] In one embodiment of this application, the compressor's operating parameters include the compressor's start-stop point temperature and start-stop cycle. The comparison of whether the compressor's operating parameters in the first operating data are the same as those in the second operating data includes:
[0027] Compare whether the compressor start-up and stop-up temperatures in the first operating data are the same as those in the second operating data;
[0028] If the compressor start-stop point temperature in the first operating data is the same as the compressor start-stop point temperature in the second operating data, then compare whether the compressor start-stop cycle in the first operating data is the same as the compressor start-stop cycle in the second operating data.
[0029] If the compressor start-stop cycle in the first operating data is the same as the compressor start-stop cycle in the second operating data, then it is determined that at least one key data in the first operating data is the same as the corresponding key data in the second operating data.
[0030] If the compressor start-stop point temperature in the first operating data is different from the compressor start-stop point temperature in the second operating data, or if the compressor start-stop cycle in the first operating data is different from the compressor start-stop cycle in the second operating data, then it is determined that at least one key data in the first operating data is different from the corresponding key data in the second operating data.
[0031] The above technical solution has the following advantages or beneficial effects:
[0032] By comparing the compressor's start-stop temperature and start-stop cycle in the first operating data with those in the second operating data, it can be determined relatively accurately whether the operating condition of the refrigeration unit to be diagnosed is the same as that of the refrigeration unit under a preset environment. Specifically, if the compressor's start-stop temperature and start-stop cycle in the first operating data are the same as those in the second operating data, then it can be determined that the operating condition of the refrigeration unit to be diagnosed is the same as that of the refrigeration unit under the preset environment. However, if at least one of the compressor's start-stop temperature and start-stop cycle in the first operating data differs from that in the second operating data, then it can be determined that the operating condition of the refrigeration unit to be diagnosed is different from that of the refrigeration unit under the preset environment. That is, the first operating data itself may be fault data, or the first operating data may be data under normal operating conditions of the refrigeration unit to be diagnosed, but the operating condition of the refrigeration unit to be diagnosed is different from that of the refrigeration unit under the preset environment.
[0033] In one embodiment of this application, performing a first correction process on the first running data based on the second running data includes:
[0034] When the compressor start-stop point temperature in the first operating data is the same as the compressor start-stop point temperature in the second operating data, but the compressor start-stop cycle in the first operating data is different from the compressor start-stop cycle in the second operating data, the first start-stop cycle of the compressor in the first operating data and the second start-stop cycle of the compressor in the second operating data are obtained.
[0035] When the first start-stop cycle is greater than the second start-stop cycle, calculate the first ratio of the first start-stop cycle to the second start-stop cycle.
[0036] The first running data is extracted and processed according to the first ratio.
[0037] The extracted data is then augmented according to the first ratio so that the compressor start-stop cycle in the first operating data is the same as the compressor start-stop cycle in the second operating data.
[0038] The above technical solution has the following advantages or beneficial effects:
[0039] If the compressor's start-stop point temperature in the first operating data is the same as that in the second operating data, but the compressor's start-stop cycle in the first operating data is different from that in the second operating data, then, if the first start-stop cycle of the compressor in the first operating data is greater than the second start-stop cycle of the compressor in the second operating data, a first ratio of the first start-stop cycle to the second start-stop cycle can be calculated. This first ratio is then used to extract and expand the first operating data, making the compressor's start-stop cycle in the first operating data the same as that in the second operating data. This ensures that the key data in the first operating data is the same as the corresponding key data in the second operating data, thus correcting the operating data of the refrigeration device to be diagnosed to be close to the operating data of the refrigeration device under a preset environment.
[0040] In one embodiment of this application, after performing a first correction process on the first running data based on the second running data, the method further includes:
[0041] If the key data in the first operating data cannot be corrected to be the same as the corresponding key data in the second operating data, then after the data collection cycle, return to the step of obtaining the first operating data of the refrigeration device to be diagnosed within the collection cycle.
[0042] The above technical solution has the following advantages or beneficial effects:
[0043] Considering that key data in the first operating data of the refrigeration device under diagnosis within a short acquisition period may not be corrected to match the corresponding key data in the second operating data, the data acquisition period can be further extended to re-acquire the first and second operating data after the extended acquisition period. This allows for a longer observation period to determine whether the operating status of the refrigeration device under diagnosis is close to that of the refrigeration device under a preset environment. Since fault diagnosis using a fault diagnosis model will be inaccurate if the operating data of the refrigeration device under diagnosis cannot be corrected to match that of the refrigeration device under a preset environment, it is unsuitable for fault diagnosis. By extending the acquisition period to assess the operating status, it is possible to avoid arbitrarily concluding that a fault diagnosis model is unusable simply because the operating data of the refrigeration device under diagnosis within a short acquisition period cannot be corrected to match that of the refrigeration device under a preset environment.
[0044] In one embodiment of this application, after the data collection period for the growth data, the method further includes:
[0045] If the key data in the first operating data cannot be corrected to be the same as the corresponding key data in the second operating data, then a clustering algorithm is used to diagnose the fault in the first operating data of the refrigeration device to be diagnosed.
[0046] The above technical solution has the following advantages or beneficial effects:
[0047] If, after the data collection period, the key data in the first operating data still cannot be corrected to match the corresponding key data in the second operating data, it indicates that the operating data of the refrigeration unit to be diagnosed cannot be corrected to be close to the operating data under the preset environment. In this case, the diagnosis results using the fault diagnosis model will be inaccurate. Therefore, clustering algorithms are required for accurate fault diagnosis.
[0048] To achieve the above objectives, embodiments of this application also propose a refrigeration device, comprising:
[0049] The box contains a storage compartment.
[0050] A data acquisition module, located inside the housing, is used to collect the operating data of the refrigeration device;
[0051] A communication module is disposed inside the housing. The communication module is electrically connected to the acquisition module and is also connected to a cloud platform. The communication module is used to send the operating data of the refrigeration device acquired by the acquisition module to the cloud platform, so that the cloud platform can execute the fault diagnosis method described in any embodiment of this application.
[0052] The above technical solution has the following advantages or beneficial effects:
[0053] The refrigeration device includes a data acquisition module and a communication module. The communication module can send the operating data of the refrigeration device collected by the acquisition module to the cloud platform. This allows the cloud platform to correct the key data in the operating data of the refrigeration device to be the same as the key data in the normal operating data of the refrigeration device under the preset environment when it determines that the operating status of the refrigeration device is different from that under the preset environment. This corrects the operating data of the refrigeration device to be closer to the operating data of the refrigeration device under the preset environment. Then, it calls the fault diagnosis model trained under the preset environment for diagnosis, thereby improving the diagnostic accuracy.
[0054] To achieve the above objectives, this application provides a fault diagnosis system for a refrigeration device, comprising:
[0055] The refrigeration device to be diagnosed;
[0056] A cloud platform is communicatively connected to the refrigeration device to be diagnosed. The cloud platform stores a fault diagnosis model. The cloud platform is used to execute the fault diagnosis method described in any embodiment of this application. The fault diagnosis model is obtained by simulating various faults of the refrigeration device under a preset environment and collecting and training data.
[0057] The above technical solution has the following advantages or beneficial effects:
[0058] The cloud platform stores fault diagnosis models obtained by simulating various faults of refrigeration devices under preset environments and collecting and training data. After obtaining the first operating data of the refrigeration device to be diagnosed, the cloud platform can first determine whether the operating status of the refrigeration device to be diagnosed is the same as that of the refrigeration device under preset environments. If it is determined that the operating status of the refrigeration device is different from that of the refrigeration device under preset environments, the key data in the operating data of the refrigeration device is corrected to be the same as the key data in the normal operating data of the refrigeration device under preset environments. This corrects the operating data of the refrigeration device to be close to the operating data of the refrigeration device under preset environments. Then, the fault diagnosis model trained under preset environments is called to perform diagnosis, thereby improving the diagnostic accuracy. Attached Figure Description
[0059] Figure 1 This is a front perspective view of a refrigeration device according to an embodiment of this application.
[0060] Figure 2 for Figure 1 A cross-sectional view.
[0061] Figure 3 This is a schematic diagram of a fault diagnosis system for a refrigeration device provided in an embodiment of this application.
[0062] Figure 4 This is a flowchart of a fault diagnosis method for a refrigeration device provided in Embodiment 1 of this application.
[0063] Figure 5 This is a flowchart illustrating the steps of comparing the compressor's operating parameters in the first operating data with those in the second operating data, as provided in an embodiment of this application.
[0064] Figure 6 This is a flowchart of the steps for performing a first correction process on first running data based on second running data, provided in an embodiment of this application.
[0065] Figure 7 This is a flowchart of a fault diagnosis method for a refrigeration device provided in Embodiment 2 of this application.
[0066] Figure label:
[0067] Container 1, refrigerator compartment 11, freezer compartment 12, compressor compartment 13, door 14, compressor 21, cloud platform 31, refrigeration device to be diagnosed 32. Detailed Implementation
[0068] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.
[0069] It should be noted that although functional modules are divided in the device schematic diagram and a logical order is shown in the flowchart, in some cases, the steps shown or described may be performed in a different order than the module division in the device or the order in the flowchart. The terms "first," "second," etc., in the specification, claims, and the aforementioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence.
[0070] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.
[0071] Currently, whether in competing refrigeration equipment manufacturers or across upstream and downstream industries, most fault diagnosis solutions employ deep learning algorithms. For refrigeration equipment fault diagnosis, a large amount of data is collected by simulating the user environment in a pre-defined setting (typically a laboratory environment). This collected data is then used to train a fault diagnosis model, resulting in a model for fault diagnosis. The trained model can be deployed on a cloud platform. When mass-produced refrigeration equipment undergoes fault diagnosis, the fault data is input into this model, and the result can be directly inferred.
[0072] However, because this fault diagnosis model is trained by simulating various faults under a preset environment, it has a high fault diagnosis rate in the preset environment, but its performance in diagnosing refrigeration units in actual user environments is not very good. One reason is that the actual user environment is different from the laboratory environment (i.e., the preset environment). Whether in the north or south, or summer or winter, there are significant differences in temperature and humidity. Another, and more important reason, is that the food stored in refrigeration units in actual user environments cannot be simulated. The amount and type of food stored in each user's refrigeration unit are different. The conditions simulated in the preset environment can only represent the usage of most users, but cannot cover all users. Therefore, when the fault diagnosis model trained under the preset environment is applied to diagnose faults in refrigeration units in actual user environments, it suffers from low diagnostic accuracy.
[0073] Based on this, this application proposes a fault diagnosis method for a refrigeration device, which aims to improve the diagnostic accuracy by correcting the operating data of the refrigeration device to be diagnosed to be close to the operating data of the refrigeration device under a preset environment and then inputting it into the fault diagnosis model for diagnosis.
[0074] The refrigeration device in this application embodiment can be a freezer, refrigerator, bar cabinet, or other refrigeration cabinet. The following uses a refrigerator as an example to describe in detail the improved technical solution of the refrigeration device in this application embodiment.
[0075] Figure 1 This is a front perspective view of a refrigeration device according to an embodiment of this application. Figure 2 for Figure 1 A cross-sectional view.
[0076] Please see Figures 1 to 2 As shown, the refrigerator provided in this embodiment may include a cabinet 1. The cabinet 1 may have a hollow structure, such as a cuboid. The cabinet 1 forms the outer shell of the refrigerator. It should be noted that the cabinet 1 may also have a hollow shell structure of other shapes.
[0077] Please see Figure 2 As shown, in some embodiments, the interior of the housing 1 may have several compartments, which may include a refrigerator compartment 11 and a freezer compartment 12. The refrigerator compartment 11 and the freezer compartment 12 may be configured as multiple storage compartments.
[0078] Please see Figure 2 As shown, in some embodiments, the refrigerator compartment 11 and the freezer compartment 12 can be used as independent storage spaces to meet different refrigeration needs such as freezing and refrigeration according to different types of food, and to store items that need to be refrigerated or frozen. The refrigerator compartment 11 and the freezer compartment 12 can be arranged vertically or horizontally.
[0079] Please see Figure 2 As shown, in some embodiments, the refrigerator may include a refrigerator liner. A refrigerator compartment 11 and a freezer compartment 12 may be formed within the refrigerator liner.
[0080] Please see Figure 2 As shown, in some embodiments, the refrigerator may include a door 14. The door 14 may be hinged to the front side of the body 1 for opening and closing the refrigerator compartment 11 and the freezer compartment 12.
[0081] It should be noted that multiple doors 14 can be provided. Each door 14 can be provided in a one-to-one correspondence with a refrigerator compartment 11 and a freezer compartment 12. One refrigerator compartment 11 can be provided with one or more doors 14, and one freezer compartment can also be provided with one or more doors 14.
[0082] Please see Figure 2As shown, in some embodiments, the refrigerator may include a refrigeration system. The refrigeration system may be located inside the refrigerator body 1. The refrigeration system can be used to provide cold air to the interior of the refrigerator to maintain a low-temperature environment in each of the refrigerator compartments 11 and the freezer compartment 12. A refrigeration system is a system that uses refrigerant circulation to lower the temperature, and mainly includes major components such as a compressor, condenser, throttling element, and evaporator. The refrigeration system achieves a cooling effect by circulating refrigerant to transfer heat from a low-temperature object to a high-temperature object.
[0083] In some embodiments, the refrigeration system may include a compressor 21. The compressor 21 can serve as a power source for the refrigeration cycle, drawing in low-temperature, low-pressure refrigerant gas and compressing it into high-temperature, high-pressure gas. The compressor can deliver the high-temperature, high-pressure refrigerant to the condenser.
[0084] In some embodiments, the refrigeration system may include a condenser. The condenser may be used to receive refrigerant flowing out of compressor 21, and may cool and convert the high-temperature, high-pressure refrigerant gas from compressor 21 into a liquid state. The condenser may transfer heat from the refrigerant to the surrounding air, thereby lowering the temperature of the refrigerant.
[0085] In some embodiments, the refrigeration system may include a throttling device (not shown). A condenser can deliver condensed refrigerant to the throttling device. The throttling device may be a capillary tube. The throttling device can be used to reduce the pressure of the refrigerant.
[0086] In some embodiments, the refrigeration system may include an evaporator (not shown). A throttling device may deliver a throttled and depressurized refrigerant into the evaporator. The evaporator may be used for the refrigerant vapor to evaporate and boil, thereby absorbing heat from the surrounding medium.
[0087] In some embodiments, the compressor, condenser, throttling device, and evaporator can be connected in sequence to form a refrigeration circuit. The refrigerant can circulate within the refrigeration circuit to cool the refrigerator compartment 11 and freezer compartment 12 inside the cabinet 1.
[0088] Please see Figure 2 As shown, in some embodiments, a compressor chamber 13 may be provided inside the housing 1. A compressor 21 may be provided inside the compressor chamber 13. The compressor chamber 13 may be located in the bottom region of the housing 1. The compressor chamber 13 may be located below the rear side of the freezer compartment 12. The compressor, condenser, throttling device, etc., may be located inside the housing 1.
[0089] It should be noted that in some other embodiments, the compressor chamber 13 may also be located at the top or side of the housing 1.
[0090] In some embodiments, a data acquisition module may also be installed inside the cabinet 1 to collect the refrigerator's operating data, including the set temperature of each compartment, the sensor temperature of each compartment, the sensor temperature of the evaporator, the speed of each fan, and the operating parameters of the compressor. A communication module may also be installed inside the cabinet 1, through which the refrigerator can communicate with a cloud platform, thereby sending the refrigerator's operating data to the cloud platform. The cloud platform, upon receiving the refrigerator's operating data, processes it and invokes a fault diagnosis model for fault diagnosis. When it is determined that the refrigerator's operating condition differs from that of a refrigerator operating under a preset environment, the key data in the refrigerator's operating data can be corrected to be the same as the key data in the refrigerator's normal operating data under the preset environment. This corrects the refrigerator's operating data to be closer to the operating data under the preset environment. Then, the fault diagnosis model trained under the preset environment is invoked for diagnosis, thereby improving the diagnostic accuracy.
[0091] Reference Figure 3 , Figure 3 This is a schematic diagram of a fault diagnosis system for a refrigeration device provided in an embodiment of this application. Figure 3 As shown, the fault diagnosis model includes a cloud platform 31 and a refrigeration device 32 to be diagnosed. The refrigeration device 32 to be diagnosed may include multiple devices, such as refrigeration device A, refrigeration device B, refrigeration device C, etc. The refrigeration device 32 to be diagnosed can be a refrigeration device corresponding to different user environments, such as a refrigeration device under user environment A, user environment B, user environment C, etc. The cloud platform 31 can store fault diagnosis models obtained by simulating various faults of the refrigeration device under preset environments and by collecting and training data. The specific training of the fault diagnosis model is as follows:
[0092] Various fault simulations are performed and data is collected under a preset environment. These faults can occur under different set temperatures and ambient temperatures. A large amount of data needs to be collected by simulating various faults under the preset environment. The data collected from the simulated faults under the preset environment is used as training samples, involving tens of thousands, hundreds of thousands, or even millions of data points. These training samples are preprocessed, and key parameters (such as the set temperature of each compartment, the sensor temperature of each compartment, the evaporator sensor temperature, the speed of each fan, and the operating parameters of the compressor) are selected and normalized. This data is then trained and continuously optimized based on a mature deep learning algorithm framework to finally generate a fault diagnosis model. After the fault diagnosis model is validated on a validation set and found to be correct, it is deployed on a cloud platform. However, in actual testing, if the fault data simulated under the preset environment is used for validation, the accuracy will be very high, but if the data actually reported by the refrigeration unit to be diagnosed is used for validation, the accuracy may be relatively poor. The reason for this is that the user's operating environment differs from the preset environment. Whether in the north or south, or summer or winter, temperature and humidity vary greatly. Furthermore, the food stored in the refrigeration unit under the user's operating environment cannot be simulated; the food and quantity stored in each refrigeration unit to be diagnosed are different. To address this, this application's embodiment improves the accuracy of the fault diagnosis model when applied to refrigeration units under the user's operating environment by correcting the operating data of the refrigeration unit to be diagnosed to be closer to the operating data of the refrigeration unit under the preset environment before inputting it into the fault diagnosis model.
[0093] Example 1
[0094] Reference Figure 4 , Figure 4 This is a flowchart of a fault diagnosis method for a refrigeration device provided in Embodiment 1 of this application. It is executed by the cloud platform 31 in the fault diagnosis system provided in any embodiment of this application, including but not limited to steps S410 to S490.
[0095] Step S410: Obtain the first operating data of the refrigeration device to be diagnosed within the acquisition period;
[0096] Step S420: Extract at least one key data from the first running data;
[0097] Step S430: Compare whether the key data is the same as the corresponding key data in the second operating data. The second operating data is the data of the refrigeration device operating normally under a preset environment. The first operating data and the second operating data have the same collection period.
[0098] Step S440: If the key data is not the same as the key data corresponding to the second running data, then the first running data is corrected according to the second running data so that the key data in the first running data is the same as the key data corresponding to the second running data.
[0099] Step S450: Call the fault diagnosis model to perform fault diagnosis on the corrected first operating data and obtain the diagnosis result. The fault diagnosis model is obtained by simulating various faults of the refrigeration unit under a preset environment and collecting and training data.
[0100] Step S460: Determine whether the diagnostic result indicates that the refrigeration unit to be diagnosed has a fault;
[0101] Step S470: If the diagnosis result indicates that the refrigeration unit to be diagnosed has a fault, then output the corresponding fault information.
[0102] Step S480: If the diagnosis result is that the refrigeration device to be diagnosed is operating normally, then the subsequent operating data of the refrigeration device to be diagnosed is corrected according to the first correction processing method, and then the fault diagnosis model is called to perform fault diagnosis on the subsequent corrected operating data of the refrigeration device to be diagnosed.
[0103] Step S490: If the key data is the same as the corresponding key data in the second operating data, then the fault diagnosis model is called to perform fault diagnosis on the subsequent operating data of the refrigeration unit to be diagnosed, and the diagnosis result is obtained.
[0104] In this embodiment, after acquiring the first operating data of the refrigeration device to be diagnosed, the cloud platform 31 first extracts at least one key data from the first operating data. The key data may include the operating parameters of the compressor, the operating parameters of the fan, and the operating parameters of the evaporator. Then, it compares whether the key data is the same as the corresponding key data in the second operating data of the refrigeration device under a preset environment. The second operating data is the data of the refrigeration device operating normally under the preset environment, and the first and second operating data have the same acquisition period. For example, it can compare whether the operating parameters of the compressor in the first operating data are the same as those in the second operating data, or whether the operating parameters of the fan in the first operating data are the same as those in the second operating data, or whether the operating parameters of the evaporator in the first operating data are the same as those in the second operating data. If the key data in the first operating data of the refrigeration device to be diagnosed is the same as the corresponding key data in the second operating data, such as the operating parameters of the compressor in the first operating data being the same as those in the second operating data, it indicates that the refrigeration device to be diagnosed is operating normally, and its operating condition is the same as that of the refrigeration device under the preset environment. At this point, fault diagnosis can be performed directly using the fault diagnosis model trained in the preset environment, and accurate diagnostic results can be obtained.
[0105] If the key data in the first operating data of the refrigeration unit to be diagnosed is different from the corresponding key data in the second operating data of the refrigeration unit under the preset environment, it indicates that the first operating data itself is fault data, meaning that the refrigeration unit to be diagnosed is currently operating with a fault. Alternatively, the first operating data may not be fault data, and the refrigeration unit to be diagnosed is currently operating normally, but its operating condition is different from that of the refrigeration unit under the preset environment. In this case, the first operating data can be further corrected based on the second operating data to make the key data in the first operating data the same as the corresponding key data in the second operating data. This process makes the first operating data of the refrigeration unit to be diagnosed closer to the second operating data of the refrigeration unit under the preset environment. Then, the fault diagnosis model trained under the preset environment can be used for fault diagnosis to obtain an accurate diagnosis result.
[0106] When the fault diagnosis model indicates that the refrigeration unit under test has a fault, meaning the first operating data is fault data, the model can directly output corresponding fault information, such as the specific fault type and fault object. When the model indicates that the refrigeration unit under test does not have a fault, meaning the first operating data is data from the unit's normal operating conditions, but the key data in the first operating data differs from the corresponding key data in the second operating data, it indicates that although the refrigeration unit is operating normally, its operating condition differs from that under the preset environment. In this case, directly using the fault diagnosis model trained under the preset environment to diagnose subsequent operating data of the refrigeration unit under test will result in inaccurate results. Therefore, to improve the accuracy of fault diagnosis, when the refrigeration unit is operating normally but its operating condition differs from that under the preset environment, the subsequent operating data of the refrigeration unit under test must be corrected according to the first correction process before the fault diagnosis model is called to diagnose the fault based on the corrected operating data. This ensures the accuracy of the fault diagnosis.
[0107] It should be noted that the acquisition period for the first and second operating data being compared must be the same to ensure the accuracy of the comparison results. For example, if the first operating data is the operating data of the refrigeration unit to be diagnosed for one day (i.e., the acquisition period is one day), then the corresponding second operating data for comparison should also be the normal operating data of the refrigeration unit under the preset environment for one day. If the first operating data is the operating data of the refrigeration unit to be diagnosed for one week (i.e., the acquisition period is one week), then the corresponding second operating data for comparison should also be the normal operating data of the refrigeration unit under the preset environment for one week.
[0108] In some embodiments, key data may include compressor operating parameters. Specifically, by extracting the compressor operating parameters from the first operating data and determining whether these parameters are identical to those in the second operating data, it can be determined whether the compressor operating state of the refrigeration device to be diagnosed is consistent with the compressor operating state of the refrigeration device under a preset environment. If the compressor operating state of the refrigeration device to be diagnosed is consistent with the compressor operating state of the refrigeration device under the preset environment, it can be generally considered that the operating condition of the refrigeration device to be diagnosed is the same as that of the refrigeration device under the preset environment.
[0109] The compressor's operating parameters may include its start-stop temperature and start-stop cycle. Generally, if the compressor's start-stop temperature and start-stop cycle are the same, then the compressor's operating state can be determined to be the same. That is, if the compressor's start-stop temperature and start-stop cycle in the first operating data are the same as those in the second operating data...
[0110] Reference Figure 5 , Figure 5 This is a flowchart of the steps for comparing the compressor operating parameters in the first operating data with the compressor operating parameters in the second operating data, provided in an embodiment of this application, including but not limited to steps S510 to S540.
[0111] Step S510: Compare whether the compressor start-stop point temperature in the first operating data is the same as the compressor start-stop point temperature in the second operating data.
[0112] Step S520: If the compressor start-stop point temperature in the first operating data is the same as the compressor start-stop point temperature in the second operating data, then compare whether the compressor start-stop cycle in the first operating data is the same as the compressor start-stop cycle in the second operating data.
[0113] Step S530: If the compressor start-stop cycle in the first operating data is the same as the compressor start-stop cycle in the second operating data, then it is determined that the key data in the first operating data is the same as the corresponding key data in the second operating data.
[0114] Step S540: If the compressor start-stop point temperature in the first operating data is different from the compressor start-stop point temperature in the second operating data, or if the compressor start-stop cycle in the first operating data is different from the compressor start-stop cycle in the second operating data, then it is determined that the key data in the first operating data is different from the corresponding key data in the second operating data.
[0115] In this embodiment, by determining whether the compressor's start-stop point temperature and start-stop cycle are the same, the operating state of the compressor can be determined relatively accurately. If the compressor operating state of the refrigeration device to be diagnosed is consistent with the compressor operating state of the refrigeration device under a preset environment, it can be determined that the operating condition of the refrigeration device to be diagnosed is basically consistent with the operating condition of the refrigeration device under the preset environment. Specifically, if the compressor's start-stop point temperature and start-stop cycle in the first operating data are the same as those in the second operating data, it can be determined that the operating condition of the refrigeration device to be diagnosed is basically the same as that of the refrigeration device under the preset environment. However, if at least one of the compressor's start-stop point temperature and start-stop cycle in the first operating data differs from that in the second operating data, it can be determined that the operating condition of the refrigeration device to be diagnosed is different from that of the refrigeration device under the preset environment.
[0116] Reference Figure 6 , Figure 6 This is a flowchart of the steps for performing a first correction process on the first running data based on the second running data, provided in an embodiment of this application, including but not limited to steps S610 to S640.
[0117] Step S610: When the compressor start-stop point temperature in the first operating data is the same as the compressor start-stop point temperature in the second operating data, but the compressor start-stop cycle in the first operating data is different from the compressor start-stop cycle in the second operating data, obtain the first start-stop cycle of the compressor in the first operating data and the second start-stop cycle of the compressor in the second operating data.
[0118] Step S620: When the first start-stop cycle is greater than the second start-stop cycle, calculate the first ratio of the first start-stop cycle to the second start-stop cycle.
[0119] Step S630: Perform data extraction processing on the first running data according to the first ratio;
[0120] In step S640, the extracted data is further expanded according to the first ratio so that the compressor start-stop cycle in the first operating data is the same as the compressor start-stop cycle in the second operating data.
[0121] In this embodiment, when the compressor's start-stop point temperature in the first operating data is the same as that in the second operating data, but the compressor's start-stop cycle in the first operating data is different from that in the second operating data, it can be determined that the operating condition of the refrigeration device to be diagnosed is different from that of the refrigeration device under a preset environment. In this case, the first operating data of the refrigeration device to be diagnosed needs to be corrected to be closer to the second operating data of the refrigeration device under the preset environment. Specifically, the first start-stop cycle of the compressor in the first operating data and the second start-stop cycle of the compressor in the second operating data can be obtained. When the first start-stop cycle is greater than the second start-stop cycle, a first ratio of the first start-stop cycle divided by the second start-stop cycle is calculated. Data extraction processing is performed on the first operating data according to the first ratio, and then the extracted data is expanded according to the first ratio. Specifically, one data point can be extracted from each data point of the first ratio in the first operating data, and the extracted data can be expanded by a multiple of the first ratio, so that the compressor's start-stop cycle in the first operating data is the same as that in the second operating data. For example, in the preset environment, the refrigeration device has a compressor that completes one start-stop cycle in one hour, while the refrigeration device to be diagnosed has a compressor that completes one start-stop cycle in two hours. That is, the first start-stop cycle is 2 hours, and the second start-stop cycle is 1 hour. At this time, the first ratio of the first start-stop cycle to the second start-stop cycle is calculated to be 2. Then, one data point is extracted from every two data points of the first operating data. Since the acquisition period of the first and second operating data is the same, the extracted data needs to be expanded by 2 times, so that the compressor start-stop cycle in the corrected first operating data is the same as the compressor start-stop cycle in the second operating data.
[0122] Example 2
[0123] Reference Figure 7 , Figure 7 This is a flowchart of a fault diagnosis method for a refrigeration device provided in Embodiment 2 of this application. It is executed by the cloud platform 31 in the fault diagnosis system provided in any embodiment of this application, including but not limited to steps S710 to S7110.
[0124] Step S710: Obtain the first operating data of the refrigeration device to be diagnosed within the acquisition period;
[0125] Step S720: Extract at least one key data from the first running data;
[0126] Step S730: Compare whether the key data is the same as the corresponding key data in the second operating data. The second operating data is the data of the refrigeration device operating normally under a preset environment. The first operating data and the second operating data have the same collection period.
[0127] Step S740: If the key data is different from the key data corresponding to the second running data, then the first running data is corrected according to the second running data.
[0128] Step S750: Determine whether the key data in the first running data can be corrected to be the same as the corresponding key data in the second running data;
[0129] Step S760: If the key data in the first operating data can be corrected to be the same as the corresponding key data in the second operating data, then the fault diagnosis model is called to perform fault diagnosis on the corrected first operating data and obtain the diagnosis result. The fault diagnosis model is obtained by simulating various faults of the refrigeration device under a preset environment and collecting and training data.
[0130] Step S770: Determine whether the diagnostic result indicates that the refrigeration unit to be diagnosed has a fault;
[0131] Step S780: If the diagnosis result indicates that the refrigeration unit to be diagnosed has a fault, then output the corresponding fault information.
[0132] Step S790: If the diagnosis result is that the refrigeration device to be diagnosed is operating normally, then the subsequent operating data of the refrigeration device to be diagnosed is corrected according to the first correction processing method, and then the fault diagnosis model is called to perform fault diagnosis on the subsequent corrected operating data of the refrigeration device to be diagnosed.
[0133] Step S7100: If the key data in the first running data cannot be corrected to be the same as the corresponding key data in the second running data, then after increasing the data collection period, return to step S710.
[0134] Step S7110: If the key data is the same as the corresponding key data in the second operating data, then the fault diagnosis model is called to perform fault diagnosis on the subsequent operating data of the refrigeration unit to be diagnosed, and the diagnosis result is obtained.
[0135] In this embodiment, when the key data in the first operating data is different from the corresponding key data in the second operating data, the first operating data needs to be corrected based on the second operating data. However, it is possible that even correcting the first operating data cannot make the key data in the first operating data the same as the corresponding key data in the second operating data. In this case, it is considered that the acquisition period of the first operating data may be too short. Therefore, the acquisition period can be extended, and the first operating data of the refrigeration device to be diagnosed can be reacquired within the acquisition period. For example, the acquisition period can be extended from 1 day to 1 week, and the first operating data of the refrigeration device to be diagnosed for 1 week can be acquired. At the same time, the key data extracted from the first operating data is compared with the corresponding key data in the second operating data of the refrigeration device under the preset environment for 1 week to determine whether the operating status of the refrigeration device to be diagnosed is the same as the operating status of the refrigeration device under the preset environment. If they are different, the first operating data is corrected based on the second operating data, and it is determined whether the key data in the first operating data can be corrected to be the same as the corresponding key data in the second operating data. Therefore, it avoids the situation where the operating data of the refrigeration device to be diagnosed cannot be corrected to be close to the operating data of the refrigeration device under the preset environment within a short collection period, and it is therefore wrong to arbitrarily conclude that the fault diagnosis model cannot be used for fault diagnosis.
[0136] In this embodiment, if, after the data collection period, the key data in the first operating data still cannot be corrected to be the same as the corresponding key data in the second operating data, it indicates that the operating data of the refrigeration device to be diagnosed cannot be corrected to be close to the operating data under the preset environment. In this case, the diagnosis result using the fault diagnosis model will be inaccurate, and a clustering algorithm is needed to perform fault diagnosis on the first operating data of the refrigeration device to be diagnosed. That is, when the operating data of the refrigeration device to be diagnosed cannot be corrected to be close to the normal operating data of the refrigeration device under the preset environment, a clustering algorithm is required for fault diagnosis.
[0137] In this embodiment, when the key data in the first operating data is the same as the corresponding key data in the second operating data, it indicates that the refrigeration device to be diagnosed is operating normally and its operating condition is the same as that of the refrigeration device under the preset environment. In this case, the fault diagnosis model trained under the preset environment can be directly called to perform fault diagnosis, and an accurate diagnosis result can be obtained. However, if the key data in the first operating data is different from the corresponding key data in the second operating data, it indicates that the refrigeration device to be diagnosed is currently faulty, or that the refrigeration device to be diagnosed is operating normally but its operating condition is different from that of the refrigeration device under the preset environment. In this case, the first operating data can be first corrected according to the second operating data, that is, the key data in the first operating data can be corrected to be the same as the corresponding key data in the second operating data, and then the fault diagnosis model can be called to perform fault diagnosis to determine whether the refrigeration device to be diagnosed is currently faulty. If a fault exists, the corresponding fault information is output. If no fault exists, it is considered that the refrigeration device to be diagnosed may subsequently experience a fault. For the subsequent operating data of the refrigeration device to be diagnosed, after the first correction processing, it is input into the fault diagnosis model for fault diagnosis, which can improve the efficiency of fault diagnosis. In this embodiment, when the key data in the first operating data is different from the corresponding key data in the second operating data, the operating data of the refrigeration device to be diagnosed can be corrected to be close to the operating data of the refrigeration device under a preset environment by correcting the key data in the first operating data to be the same as the corresponding key data in the second operating data. Then, the fault diagnosis model can be called to perform fault diagnosis, which can improve the diagnostic accuracy of the fault diagnosis model applied to the refrigeration device under the user's operating environment.
[0138] The embodiments described in this application are for the purpose of more clearly illustrating the technical solutions of the embodiments of this application, and do not constitute a limitation on the technical solutions provided by the embodiments of this application. As those skilled in the art will know, with the evolution of technology and the emergence of new application scenarios, the technical solutions provided by the embodiments of this application are also applicable to similar technical problems.
[0139] Those skilled in the art will understand that the technical solutions shown in the figures do not constitute a limitation on the embodiments of this application, and may include more or fewer steps than shown, or combine certain steps, or different steps.
[0140] Those skilled in the art will understand that all or some of the steps in the methods disclosed above, as well as the functional modules / units in the systems and devices, can be implemented as software, firmware, hardware, or suitable combinations thereof.
[0141] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms “comprising” and “having,” and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.
[0142] It should be understood that in this application, "at least one (item)" means one or more, and "more than" means two or more. "And / or" is used to describe the relationship between related objects, indicating that three relationships can exist. For example, "A and / or B" can represent three cases: only A exists, only B exists, and both A and B exist simultaneously, where A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one (item) of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one (item) of a, b, or c can represent: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be single or multiple.
[0143] The preferred embodiments of the present application have been described above with reference to the accompanying drawings, but this does not limit the scope of the claims of the present application. Any modifications, equivalent substitutions, and improvements made by those skilled in the art without departing from the scope and substance of the embodiments of the present application shall be within the scope of the claims of the present application.
Claims
1. A fault diagnosis method for a refrigeration device, characterized in that, The method includes: Acquire the first operating data of the refrigeration device to be diagnosed within the acquisition period; At least one key data is extracted from the first operating data, and the key data is compared with the corresponding key data in the second operating data. The second operating data is the data of the refrigeration device operating normally under a preset environment. The first operating data and the second operating data have the same collection period. If the key data is different from the key data corresponding to the second running data, then the first running data is corrected according to the second running data so that the key data in the first running data is the same as the key data corresponding to the second running data. The fault diagnosis model is invoked to diagnose the faults in the corrected first operating data and obtain the diagnosis results. The fault diagnosis model is obtained by simulating various faults of the refrigeration device under the preset environment and collecting and training data.
2. The method according to claim 1, characterized in that, After calling the fault diagnosis model to perform fault diagnosis on the corrected first operating data and obtaining the diagnosis result, the method further includes: If the diagnostic result indicates that the refrigeration device to be diagnosed has a malfunction, then the corresponding fault information is output. If the diagnostic result indicates that the refrigeration device to be diagnosed is operating normally, then the subsequent operating data of the refrigeration device to be diagnosed is corrected according to the first correction processing method, and then the fault diagnosis model is called to perform fault diagnosis on the subsequent corrected operating data of the refrigeration device to be diagnosed.
3. The method according to claim 1, characterized in that, After comparing whether the key data is the same as the corresponding key data in the second running data, the method further includes: If the key data is the same as the key data in the second operating data, then the fault diagnosis model is invoked to perform fault diagnosis on the subsequent operating data of the refrigeration device to be diagnosed, and a diagnosis result is obtained.
4. The method according to claim 1, characterized in that, The key data includes the compressor's operating parameters. Correspondingly, comparing whether the key data is the same as the key data in the second operating data includes: Compare the compressor operating parameters in the first operating data with those in the second operating data to see if they are the same.
5. The method according to claim 4, characterized in that, The compressor's operating parameters include the compressor's start-stop point temperature and start-stop cycle. The comparison of whether the compressor's operating parameters in the first operating data are the same as those in the second operating data includes: Compare whether the compressor start-up and stop-up temperatures in the first operating data are the same as those in the second operating data; If the compressor start-stop point temperature in the first operating data is the same as the compressor start-stop point temperature in the second operating data, then compare whether the compressor start-stop cycle in the first operating data is the same as the compressor start-stop cycle in the second operating data. If the compressor start-stop cycle in the first operating data is the same as the compressor start-stop cycle in the second operating data, then it is determined that at least one key data in the first operating data is the same as the corresponding key data in the second operating data. If the compressor start-stop point temperature in the first operating data is different from the compressor start-stop point temperature in the second operating data, or if the compressor start-stop cycle in the first operating data is different from the compressor start-stop cycle in the second operating data, then it is determined that at least one key data in the first operating data is different from the corresponding key data in the second operating data.
6. The method according to claim 5, characterized in that, The first correction process for the first operating data based on the second operating data includes: When the compressor start-stop point temperature in the first operating data is the same as the compressor start-stop point temperature in the second operating data, but the compressor start-stop cycle in the first operating data is different from the compressor start-stop cycle in the second operating data, the first start-stop cycle of the compressor in the first operating data and the second start-stop cycle of the compressor in the second operating data are obtained. When the first start-stop cycle is greater than the second start-stop cycle, calculate the first ratio of the first start-stop cycle to the second start-stop cycle. The first running data is extracted and processed according to the first ratio. The extracted data is then augmented according to the first ratio so that the compressor start-stop cycle in the first operating data is the same as the compressor start-stop cycle in the second operating data.
7. The method according to claim 1, characterized in that, After performing a first correction process on the first running data based on the second running data, the method further includes: If the key data in the first operating data cannot be corrected to be the same as the corresponding key data in the second operating data, then after the data collection cycle, return to the step of obtaining the first operating data of the refrigeration device to be diagnosed within the collection cycle.
8. The method according to claim 7, characterized in that, After the data collection period for the growth data, the method further includes: If the key data in the first operating data cannot be corrected to be the same as the corresponding key data in the second operating data, then a clustering algorithm is used to diagnose the fault in the first operating data of the refrigeration device to be diagnosed.
9. A refrigeration device, characterized in that, include: The box contains a storage compartment. A data acquisition module, located inside the housing, is used to collect the operating data of the refrigeration device; A communication module is disposed inside the housing. The communication module is electrically connected to the acquisition module and is also connected to a cloud platform. The communication module is used to send the operating data of the refrigeration device acquired by the acquisition module to the cloud platform so that the cloud platform can execute the method described in any one of claims 1-8.
10. A fault diagnosis system for a refrigeration device, characterized in that, include: The refrigeration device to be diagnosed; A cloud platform is communicatively connected to the refrigeration device to be diagnosed. The cloud platform stores a fault diagnosis model. The cloud platform is used to execute the method described in any one of claims 1-8, wherein the fault diagnosis model is obtained by simulating various faults of the refrigeration device under a preset environment and collecting and training data.