Fault diagnosis method and device for household equipment, and electronic equipment
The pre-built diagnostic model and data processing module handle data abnormalities in home equipment in smart home systems, and the misjudgment problems caused by data scarcity and imbalance are solved, and the accuracy of fault diagnosis is improved.
Patent Information
- Application Number
- CN202411591585.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-11-08
- Publication Date
- 2025-01-10
AI Technical Summary
There are misjudgment problems caused by scarce or unbalanced data parameters in smart home systems, which affects the accuracy of fault diagnosis.
The pre-constructed first diagnostic model and the second diagnostic model are adopted to determine the target diagnostic model based on the data of the equipment operation parameters, and the abnormal data is processed through the data expansion module and the data balance module to improve the accuracy of fault diagnosis.
Even in the case of abnormal data, home equipment fault diagnosis can be carried out more accurately, improving the accuracy of fault diagnosis in smart home systems.
Smart Images
Figure CN119292238A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of smart home and intelligent family technology, and more specifically, to a fault diagnosis method and device for household appliances, and an electronic device. Background Art
[0002] At present, smart home systems have been widely used in various aspects such as home security, environmental control, and health management. Therefore, during the operation of the smart home system, it is very necessary to perform fault diagnosis on the home appliances in the smart home system.
[0003] In the related art, the collected equipment operation parameters of the home appliances are usually analyzed based on machine learning classification algorithms (for example, decision trees, support vector machines, etc.) to determine whether there are any faults in the operation of the home appliances.
[0004] In the process of implementing the embodiments of the present disclosure, it is found that there are at least the following problems in the related art:
[0005] The collected equipment operation parameters often have data anomalies such as scarcity of one parameter or imbalanced data volume of multiple parameters. In this case, it is very easy to misjudge home equipment failures when analyzing equipment operation parameters based on machine learning classification algorithms. Therefore, how to improve the accuracy of fault diagnosis of home equipment in smart home systems has become a technical problem that needs to be solved urgently.
[0006] It should be noted that the information disclosed in the above background technology section is only used to enhance the understanding of the background of the present application, and therefore may include information that does not constitute the prior art known to ordinary technicians in the field. Summary of the invention
[0007] In order to provide a basic understanding of some aspects of the disclosed embodiments, a brief summary is given below. The summary is not an extensive review, nor is it intended to identify key / critical components or delineate the scope of protection of these embodiments, but rather serves as a prelude to the detailed description that follows.
[0008] The disclosed embodiments provide a method and device for diagnosing faults in household appliances, and an electronic device, which can improve the accuracy of diagnosing faults in household appliances in a smart home system. In some embodiments, the method for diagnosing faults in household appliances includes: obtaining equipment operating parameters of household appliances; determining the data status of the equipment operating parameters; determining a target diagnostic model in a first diagnostic model and a second diagnostic model according to the data status of the equipment operating parameters; inputting the equipment operating parameters into the target diagnostic model for processing to obtain a fault diagnosis result; wherein the first diagnostic model is a pre-constructed model that can process data anomalies of equipment operating parameters and perform fault diagnosis based on the processed equipment operating parameters; and the second diagnostic model is a pre-constructed model that can perform fault diagnosis based on equipment operating parameters with normal data.
[0009] Optionally, a target diagnostic model in the first diagnostic model and the second diagnostic model is determined according to the data situation of the equipment operating parameters, including: when there is a data anomaly in the equipment operating parameters, the first diagnostic model is determined as the target diagnostic model; when there is no data anomaly in the equipment operating parameters, the second diagnostic model is determined as the target diagnostic model.
[0010] Optionally, determining whether the equipment operating parameters comply with the data conditions includes: when any one of the equipment operating parameters is scarce or the data amount of multiple parameters is unbalanced, determining that there is a data anomaly in the equipment operating parameters; and when the equipment operating parameters include all parameters and the data amount of each parameter is balanced, determining that there is no data anomaly in the equipment operating parameters.
[0011] Optionally, a data expansion module and a fault diagnosis module are deployed in the first diagnostic model. When the first diagnostic model is the target diagnostic model, the equipment operating parameters are input into the target diagnostic model for processing to obtain a fault diagnosis result, including: when the equipment operating parameters are abnormal due to the scarcity of any parameter, the equipment operating parameters are expanded by the data expansion module; and the expanded equipment operating parameters are processed by the fault diagnosis module to obtain a fault diagnosis result.
[0012] Optionally, a data balancing module and a fault diagnosis module are deployed in the first diagnostic model. When the first diagnostic model is a target diagnostic model, the equipment operating parameters are input into the target diagnostic model for processing to obtain fault diagnosis results, including: when the equipment operating parameters are abnormal due to unbalanced data volume of multiple parameters, the fault diagnosis module adjusts the loss weights of each parameter through the data balancing module; and processes the equipment operating parameters through the fault diagnosis module after adjusting the loss weight to obtain fault diagnosis results.
[0013] Optionally, the data balancing module is used to adjust the loss weight of each parameter referred to by the fault diagnosis module, including: for each parameter with a small amount of data, increasing the loss weight of the parameter referred to by the fault diagnosis module.
[0014] Optionally, before determining the data status of the equipment operating parameters, the fault diagnosis method further includes: preprocessing the equipment operating parameters to remove noise data in the equipment operating parameters and unifying the equipment operating parameters under the same dimension.
[0015] Optionally, the first diagnostic model or the second diagnostic model is trained in the following manner: a labeled training sample set is obtained, the training sample set including a training set, a validation set and a test set; based on a preset optimization algorithm, in each iteration, a batch of data is randomly selected from the training set to train the basic model to obtain the first diagnostic model or the second diagnostic model; wherein, during the training process, the performance of the basic model is continuously tested based on the sample data in the test set, and the training of the basic model is stopped when the performance of the basic model remains unchanged within a set number of iterations; during the training process, the performance of the basic model is continuously tested based on the sample data in the validation set, and the hyperparameters of the basic model are dynamically adjusted according to the performance test results.
[0016] In some embodiments, a fault diagnosis device for household appliances includes: a fault diagnosis device for household appliances, characterized in that it includes: an acquisition module, configured to acquire equipment operating parameters of household appliances; a judgment module, configured to judge the data status of the equipment operating parameters; a determination module, configured to determine a target diagnostic model in a first diagnostic model and a second diagnostic model according to the data status of the equipment operating parameters; a processing module, configured to input the equipment operating parameters into the target diagnostic model for processing to obtain a fault diagnosis result; wherein the first diagnostic model is a pre-constructed model that can process data anomalies of equipment operating parameters and perform fault diagnosis based on the processed equipment operating parameters; the second diagnostic model is a pre-constructed model that can perform fault diagnosis based on equipment operating parameters with normal data.
[0017] In some embodiments, a fault diagnosis device for household appliances includes a processor and a memory storing program instructions, and the processor is configured to execute the fault diagnosis method for household appliances as described above when running the program instructions.
[0018] In some embodiments, an electronic device includes: a device body; and a fault diagnosis device for household appliances as described above, installed on the device body.
[0019] The fault diagnosis method and device for household appliances and the electronic device provided by the embodiments of the present disclosure can achieve the following technical effects:
[0020] The disclosed embodiment pre-constructs a first fault diagnosis model that can process data anomalies of device operation parameters and can perform fault diagnosis on home appliances based on the processed device operation parameters. A second fault diagnosis model that can directly perform fault diagnosis on home appliances based on device operation parameters without data anomalies is pre-constructed. In the disclosed embodiment, after obtaining the device operation parameters of the home appliance, first, the data situation of the device operation parameters will be confirmed. Then, the target diagnosis model in the first fault diagnosis model and the second fault diagnosis model will be selected according to the data situation of the device operation parameters. Finally, the home appliance will input the device operation parameters into the target diagnosis model to perform fault diagnosis on the home appliance to obtain the fault diagnosis result. In this way, even if the collected device operation parameters have data anomalies such as scarcity of a certain parameter or imbalance of data volume of multiple parameters, the home appliance can be more accurately diagnosed. Therefore, the disclosed embodiment can improve the accuracy of fault diagnosis of home appliances in the smart home system.
[0021] The above general description and the following description are exemplary and explanatory only and are not intended to limit the present application. BRIEF DESCRIPTION OF THE DRAWINGS
[0022] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.
[0023] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.
[0024] Figure 1 is a schematic diagram of a hardware environment for a fault diagnosis method for household appliances provided by an embodiment of the present disclosure;
[0025] Figure 2 is a schematic diagram of an electronic device provided by an embodiment of the present disclosure;
[0026] Figure 3 is a schematic diagram of a fault diagnosis method for household appliances provided by an embodiment of the present disclosure;
[0027] Figure 4 is a schematic diagram of another fault diagnosis method for household equipment provided by an embodiment of the present disclosure;
[0028] Figure 5 is a schematic diagram of a fault diagnosis device for household appliances provided by an embodiment of the present disclosure;
[0029] Figure 6 It is a schematic diagram of another fault diagnosis method and device for household appliances provided by an embodiment of the present disclosure. DETAILED DESCRIPTION
[0030] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present application.
[0031] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.
[0032] According to one aspect of the embodiments of the present disclosure, a fault diagnosis method for household appliances is provided. The fault diagnosis method can be widely used to diagnose faults of household appliances in a whole-house intelligent digital control system such as a smart home system, an intelligent home system, an intelligent home appliance ecosystem, and an intelligence house ecosystem.
[0033] Optionally, the household appliances include: smart air conditioners, smart range hoods, smart refrigerators, smart ovens, smart stoves, smart washing machines, smart water heaters, smart washing equipment, smart dishwashers, smart projection equipment, smart TVs, smart clothes drying racks, smart curtains, smart audio and video, smart sockets, smart speakers, smart speakers, smart fresh air equipment, smart kitchen and bathroom equipment, smart bathroom equipment, smart sweeping robots, smart window cleaning robots, smart mopping robots, smart air purification equipment, smart steamers, smart microwave ovens, smart kitchen treasures, smart purifiers, smart water dispensers and smart door locks, but are not limited to these.
[0034] In some embodiments, the fault diagnosis method for household equipment can be applied to Figure 1The hardware environment is composed of the electronic device 100 and a plurality of household appliances 200 (for example, a refrigerator, an air conditioner and a washing machine). Figure 1 As shown, the electronic device 100 is connected to the home appliance 200 via a network, so that the electronic device 100 can perform fault diagnosis on the home appliance 200 .
[0035] The above network may include but is not limited to at least one of the following: wired network, wireless network. The above wired network may include but is not limited to at least one of the following: wide area network, metropolitan area network, local area network, and the above wireless network may include but is not limited to at least one of the following: WIFI (Wireless Fidelity), Bluetooth.
[0036] In some embodiments, the electronic device 100 may Figure 2 As shown, it includes a device body 110 and a fault diagnosis device 500 (600) for household appliances.
[0037] Specifically, the fault diagnosis device 500 ( 600 ) for a household appliance is installed on the appliance body 110 .
[0038] Optionally, the fault diagnosis device 600 for household appliances includes a processor, which can obtain the equipment operation parameters of the household appliances and determine the data status of the equipment operation parameters. After determining the data status of the equipment operation parameters, the processor can select a target diagnosis model from the first diagnosis model and the second diagnosis model according to the data status of the equipment operation parameters. The equipment operation parameters can be input into the fault diagnosis model for processing to obtain a fault diagnosis result; wherein the first diagnosis model is a pre-constructed model that can process data anomalies of the equipment operation parameters and perform fault diagnosis based on the processed equipment operation parameters; the second diagnosis model is a pre-constructed model that can perform fault diagnosis based on equipment operation parameters with normal data.
[0039] In combination with the above electronic device, the present disclosure provides a fault diagnosis method for household equipment, such as Figure 3 As shown, the fault diagnosis method includes:
[0040] S301: The processor obtains device operating parameters of the home device.
[0041] Specifically, each home appliance in the smart home system is provided with sensors for collecting various parameters (e.g., temperature sensor, humidity sensor, voltage sensor, current sensor, etc.), and the processor of the electronic device is connected to the communication interface of these sensors through the network. Therefore, the processor can obtain the device operation parameters of the home appliance through these sensors.
[0042] S302: The processor determines the data status of the equipment operation parameters.
[0043] Specifically, by judging the data status of the device operating parameters, it can be determined whether there is scarcity in the collected device operating parameters and whether the data volume of each parameter is balanced. Thus, it can be determined whether the household device can be accurately diagnosed based on the collected device operating parameters. Therefore, after obtaining the device operating parameters of the household device, the processor needs to judge the data status of the device operating parameters.
[0044] Optionally, determining whether the equipment operating parameters comply with the data conditions includes: when any one of the equipment operating parameters is scarce or the data amount of multiple parameters is unbalanced, determining that there is a data anomaly in the equipment operating parameters; and when the equipment operating parameters include all parameters and the data amount of each parameter is balanced, determining that there is no data anomaly in the equipment operating parameters.
[0045] Specifically, if any of the device operating parameters is scarce, or the data volume of multiple parameters is unbalanced, it indicates that the reliability of the fault diagnosis result of the home appliance analyzed based on the current device operating parameters is low. Therefore, in this case, it can be determined that there is data anomaly in the device operating parameters.
[0046] Specifically, if the device operation parameters include all parameters and the data volume of each parameter is balanced, it means that the reliability of the fault diagnosis result of the home appliance analyzed based on the current device operation parameters is high. Therefore, in this case, it can be determined that there is no data anomaly in the device operation parameters.
[0047] S303: The processor determines a target diagnostic model in the first diagnostic model and the second diagnostic model according to data of the equipment operation parameters.
[0048] Among them, the first diagnostic model is a pre-built model that can process data anomalies of equipment operating parameters and perform fault diagnosis based on the processed equipment operating parameters; the second diagnostic model is a pre-built model that can perform fault diagnosis based on equipment operating parameters with normal data.
[0049] Specifically, the first diagnostic model is a pre-constructed model that can process data anomalies of device operating parameters, and after processing the data anomalies, can perform fault diagnosis on the home appliance based on the processed device operating parameters. The second diagnostic model is a pre-constructed model that can directly perform fault diagnosis on the home appliance based on the device operating parameters without data anomalies.
[0050] Optionally, a target diagnostic model in the first diagnostic model and the second diagnostic model is determined according to the data situation of the equipment operating parameters, including: when there is a data anomaly in the equipment operating parameters, the first diagnostic model is determined as the target diagnostic model; when there is no data anomaly in the equipment operating parameters, the second diagnostic model is determined as the target diagnostic model.
[0051] Specifically, if there are data anomalies in the device operating parameters, it means that the existing machine learning classification algorithm is used for fault diagnosis based on the device operating parameters, which is very likely to misjudge the home appliance fault. Therefore, in this case, the first diagnostic model needs to be determined as the target fault diagnosis model.
[0052] Specifically, if there is no data anomaly in the device operating parameters, it means that there is no scarcity in the parameters of the device operating parameters, and the data volume of each parameter is relatively balanced, and the existing machine learning classification algorithm can be used to diagnose the fault of the home appliance. Therefore, in this case, the second diagnostic model can be determined as the target diagnostic model.
[0053] Optionally, a data expansion module and a fault diagnosis module are deployed in the first diagnostic model. When the first diagnostic model is the target diagnostic model, the equipment operating parameters are input into the target diagnostic model for processing to obtain a fault diagnosis result, including: when the equipment operating parameters are abnormal due to the scarcity of any parameter, the equipment operating parameters are expanded by the data expansion module; and the expanded equipment operating parameters are processed by the fault diagnosis module to obtain a fault diagnosis result.
[0054] Specifically, when the device operating parameters are judged to be data anomalies due to the scarcity of a certain parameter, the first diagnostic model can first be expanded through the data expansion module deployed therein to compensate for the scarcity of the parameter, and then the expanded device operating parameters can be processed through the fault diagnosis module deployed therein to perform fault diagnosis on the home equipment.
[0055] Optionally, the data expansion module may use a random walk algorithm or a heat diffusion graph algorithm to expand the equipment operating parameters of scarce items.
[0056] Specifically, when the heat diffusion graph algorithm is used for expansion, taking the adjacency matrix A of the original graph of the equipment operation parameters of the scarce items as an example, the adjacency matrix A can be expanded by the following expression:
[0057]
[0058] Among them, A is the adjacency matrix of the original graph of equipment operating parameters of scarce items, D is the expanded matrix, and β is the diffusion coefficient.
[0059] For example, in the adjacency matrix of the original graph When the diffusion coefficient β = 0.8, the expanded matrix
[0060] It should be noted that when calculating the expanded matrix D, since the high-order terms gradually approach 0, it is usually sufficient to calculate only the first few terms.
[0061] Optionally, a data balancing module and a fault diagnosis module are deployed in the first diagnostic model. When the first diagnostic model is a target diagnostic model, the equipment operating parameters are input into the target diagnostic model for processing to obtain fault diagnosis results, including: when the equipment operating parameters are abnormal due to unbalanced data volume of multiple parameters, the fault diagnosis module adjusts the loss weights of each parameter through the data balancing module; and processes the equipment operating parameters through the fault diagnosis module after adjusting the loss weight to obtain fault diagnosis results.
[0062] Specifically, when the device operation parameters are judged to be data abnormal due to the imbalance of data volume of multiple parameters, the first diagnostic model can first adjust the loss weights of various parameters referred to by the fault diagnosis module when performing fault diagnosis through the data balancing module deployed therein, so as to balance the attention of the fault diagnosis module to different types of faults, and then process the device operation parameters through the fault diagnosis module after adjusting the loss weights to perform fault diagnosis on the home equipment.
[0063] Optionally, the data balancing module is used to adjust the loss weight of each parameter referred to by the fault diagnosis module, including: for each parameter with a small amount of data, increasing the loss weight of the parameter referred to by the fault diagnosis module.
[0064] Specifically, each parameter in the equipment operation parameters can be used to judge a type of fault. Therefore, for each parameter with a small amount of data, by increasing the loss weight of the fault diagnosis module referring to the parameter, the accuracy of the fault diagnosis of a type can be improved.
[0065] Specifically, the loss function of the fault diagnosis module is expressed as follows:
[0066]
[0067] Where N is the number of equipment operating parameters, y i is the i-th fault category, Output y for the i-th equipment operating parameter i The probability of For category y i The loss weight.
[0068] Exemplarily, taking the case where the data volume of the first equipment operating parameter is 100 and the data volume of the second equipment operating parameter is 20, the loss weight of the fault diagnosis module referring to the second equipment operating parameter can be increased.
[0069] Specifically, a fault diagnosis module is also deployed in the second diagnostic model. The second diagnostic model processes the device operating parameters through the fault diagnosis module to determine the fault diagnosis result of the home appliance.
[0070] S304, the processor inputs the equipment operation parameters into the target diagnosis model for processing to obtain a fault diagnosis result.
[0071] Specifically, a fault diagnosis module is also deployed in the second diagnostic model. The second diagnostic model processes the device operating parameters through the fault diagnosis module to determine the fault diagnosis result of the home appliance.
[0072] The disclosed embodiment pre-constructs a first fault diagnosis model that can process data anomalies of device operation parameters and can perform fault diagnosis on home appliances based on the processed device operation parameters. A second fault diagnosis model that can directly perform fault diagnosis on home appliances based on device operation parameters without data anomalies is pre-constructed. In the disclosed embodiment, after obtaining the device operation parameters of the home appliance, first, the data situation of the device operation parameters will be confirmed. Then, the target diagnosis model in the first fault diagnosis model and the second fault diagnosis model will be selected according to the data situation of the device operation parameters. Finally, the home appliance will input the device operation parameters into the target diagnosis model to perform fault diagnosis on the home appliance to obtain the fault diagnosis result. In this way, even if the collected device operation parameters have data anomalies such as scarcity of a certain parameter or imbalance of data volume of multiple parameters, the home appliance can be more accurately diagnosed. Therefore, the disclosed embodiment can improve the accuracy of fault diagnosis of home appliances in the smart home system.
[0073] It should be noted that when the equipment operating parameters are processed by the fault diagnosis module of the first diagnostic model and the second diagnostic model, each equipment operating parameter needs to be regarded as a node, and the data of the node and its adjacent nodes need to be statistically analyzed to calculate statistical quantities such as the mean and variance. At the same time, the consistency of the node data is judged based on the set threshold, the collaborative working situation of the various components of the home appliance is judged, and it is judged whether there is a fault in the home appliance.
[0074] Specifically, the consistency of the data is determined according to the following expression:
[0075]
[0076] Among them, x i is the data of the i-th node, and the set of its adjacent node data is is the mean, the square of s is the variance, and k is the adjustment coefficient.
[0077] In some embodiments, a nonlinear activation function is also introduced into the fault diagnosis module. In this way, the accuracy of diagnosing whether a household appliance has a fault based on complex device operating parameters can be improved by combining the above calculated consistency level.
[0078] The present disclosure provides another method for diagnosing a fault in a household appliance. Figure 4 As shown, the fault diagnosis method includes:
[0079] S401: The processor obtains device operating parameters of the home device.
[0080] S402: The processor pre-processes the device operating parameters to remove noise data in the device operating parameters and unify the device operating parameters under the same dimension.
[0081] Specifically, the steps of preprocessing the equipment operating parameters include but are not limited to: removing duplicate, erroneous and abnormal data in the equipment operating parameters, and normalizing the equipment operating parameters to unify data of different dimensions into the same range for subsequent processing.
[0082] S403: The processor determines the data status of the equipment operation parameters.
[0083] S404: The processor determines a target diagnostic model in the first diagnostic model and the second diagnostic model according to data of the equipment operation parameters.
[0084] Among them, the first diagnostic model is a pre-built model that can process data anomalies of equipment operating parameters and perform fault diagnosis based on the processed equipment operating parameters; the second diagnostic model is a pre-built model that can perform fault diagnosis based on equipment operating parameters with normal data.
[0085] S405, the processor inputs the equipment operation parameters into the target diagnosis model for processing to obtain a fault diagnosis result.
[0086] In the disclosed embodiment, before judging the data of the device operating parameters, the device operating parameters are preprocessed to remove the noise data in the device operating parameters and unify the device operating parameters to a unified dimension. In this way, the accuracy of subsequent judgment of the data situation and the accuracy of fault diagnosis of home appliances based on the device operating parameters can be improved.
[0087] In some embodiments, the first diagnostic model or the second diagnostic model is trained as follows: a labeled training sample set is obtained, and the training sample set includes a training set, a validation set and a test set; based on a preset optimization algorithm, in each iteration, a batch of data is randomly selected from the training set to train the basic model to obtain the first diagnostic model or the second diagnostic model; wherein, during the training process, the performance of the basic model is continuously tested based on the sample data in the test set, and the training of the basic model is stopped when the performance of the basic model remains unchanged within a set number of iterations; during the training process, the performance of the basic model is continuously tested based on the sample data in the validation set, and the hyperparameters of the basic model are dynamically adjusted according to the performance test results.
[0088] Specifically, the division ratios of the training set, validation set, and test set are 70%, 20%, and 10%, respectively.
[0089] Specifically, the preset optimization algorithm may be stochastic gradient descent (SGD), Adagrad, Adadelta, etc.
[0090] Specifically, during the training process, the performance of the basic model is tested, including accuracy, recall, etc. Specifically, if the performance of the basic model remains unchanged within the set number of iterations, it means that the performance of the basic model will not be improved. Therefore, the training of the basic model can be stopped to avoid the basic model from overfitting the training data.
[0091] Optionally, regularization techniques can be applied to the base model by adding the square sum of the base model parameters to the loss function and multiplying it by a regularization coefficient. This can make the base model tend to choose smaller parameter values during the learning process, reduce the complexity of the base model, and prevent overfitting.
[0092] Specifically, the hyperparameters of the basic model are dynamically adjusted based on the performance test results of the validation set on the basic model. For example, different learning rate values (such as 0.01, 0.001, 0.0001, etc.) are tried through methods such as grid search or random search, the performance of the basic model on the validation set is observed, and the optimal learning rate is selected. By experimenting with different combinations, the best balance between the diagnostic accuracy and model complexity of the basic model can be found to prevent overfitting.
[0093] In the embodiment of the present disclosure, in the process of training the basic model to obtain the first diagnostic model and the second diagnostic model, it is determined whether to stop training based on the performance test results of the test set on the basic model, and the hyperparameters of the basic model are adjusted based on the performance test results of the test set on the basic model. This can prevent overfitting of the basic model, which is conducive to improving the accuracy of the first diagnostic model and the second diagnostic model in diagnosing household equipment faults.
[0094] Combination Figure 5 As shown, the embodiment of the present disclosure provides a fault diagnosis device 500 for household appliances, including: an acquisition module 501, a judgment module 502, a determination module 503 and a processing module 504. The acquisition module 501 is configured to acquire the equipment operation parameters of the household appliances. The judgment module 502 is configured to judge the data status of the equipment operation parameters. The determination module 503 is configured to determine the target diagnosis model in the first diagnosis model and the second diagnosis model according to the data status of the equipment operation parameters. The processing module 504 is configured to input the equipment operation parameters into the target diagnosis model for processing to obtain a fault diagnosis result. Among them, the first diagnosis model is a pre-constructed model that can process data anomalies of equipment operation parameters and perform fault diagnosis according to the processed equipment operation parameters; the second diagnosis model is a pre-constructed model that can perform fault diagnosis according to equipment operation parameters with normal data.
[0095] Combination Figure 6 As shown, the embodiment of the present disclosure provides a fault diagnosis device 600 for household appliances, including: a processor (processor) 601 and a memory (memory) 602. Optionally, the fault diagnosis device 600 for household appliances may also include a communication interface (Communication Interface) 603 and a bus 604. Among them, the processor 601, the communication interface 603, and the memory 602 can communicate with each other through the bus 604. The communication interface 603 can be used for information transmission. The processor 601 can call the logic instructions in the memory 602 to execute the fault diagnosis method for household appliances in the above embodiment.
[0096] In addition, the logic instructions in the memory 602 described above may be implemented in the form of software functional units and when sold or used as independent products, may be stored in a computer-readable storage medium.
[0097] The memory 602 is a computer-readable storage medium that can be used to store software programs and computer executable programs, such as program instructions / modules corresponding to the method in the embodiment of the present disclosure. The processor 601 executes the function application and data processing by running the program instructions / modules stored in the memory 602, that is, the fault diagnosis method for home appliances in the above embodiment is implemented.
[0098] The memory 602 may include a program storage area and a data storage area, wherein the program storage area may store an operating system and an application required for at least one function; the data storage area may store data created according to the use of the terminal device, etc. In addition, the memory 602 may include a high-speed random access memory and may also include a non-volatile memory.
[0099] An embodiment of the present disclosure provides a computer-readable storage medium storing computer-executable instructions, wherein the computer-executable instructions are configured to execute the above-mentioned fault diagnosis method for household appliances.
[0100] The above is only a preferred implementation of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.
Claims
1. A fault diagnosis method for household equipment, characterized in that: include: Obtain equipment operating parameters of home equipment; Determine the data status of equipment operating parameters; Determine a target diagnostic model in the first diagnostic model and the second diagnostic model according to data of the equipment operation parameters; Input the equipment operation parameters into the target diagnosis model for processing to obtain the fault diagnosis results; Among them, the first diagnostic model is a pre-built model that can process data anomalies of equipment operating parameters and perform fault diagnosis based on the processed equipment operating parameters; the second diagnostic model is a pre-built model that can perform fault diagnosis based on equipment operating parameters with normal data.
2. The fault diagnosis method according to claim 1, characterized in that: Determining a target diagnostic model in the first diagnostic model and the second diagnostic model according to data of the equipment operation parameters includes: In the case where there is data anomaly in the equipment operation parameters, the first diagnostic model is determined as the target diagnostic model; When there is no data anomaly in the equipment operating parameters, the second diagnostic model is determined as the target diagnostic model.
3. The fault diagnosis method for household appliances according to claim 1, characterized in that: Data to determine whether the equipment operating parameters are in compliance, including: When any parameter among the equipment operation parameters is scarce or the data volume of multiple parameters is unbalanced, it is determined that there is data anomaly in the equipment operation parameters; When the equipment operation parameters include all parameters and the data volume of each parameter is balanced, it is determined that there is no data anomaly in the equipment operation parameters.
4. The fault diagnosis method for household appliances according to claim 3, characterized in that: The first diagnostic model is deployed with a data expansion module and a fault diagnosis module. When the first diagnostic model is a target diagnostic model, the equipment operation parameters are input into the target diagnostic model for processing to obtain a fault diagnosis result, including: In the case that the equipment operation parameters are abnormal due to the scarcity of any parameter, the equipment operation parameters are expanded through the data expansion module; The expanded equipment operating parameters are processed through the fault diagnosis module to obtain the fault diagnosis result.
5. The fault diagnosis method for household appliances according to claim 3, characterized in that: The first diagnostic model is deployed with a data balancing module and a fault diagnosis module. When the first diagnostic model is a target diagnostic model, the equipment operation parameters are input into the target diagnostic model for processing to obtain the fault diagnosis results, including: In the case where the equipment operation parameters are abnormal due to the imbalance of data volume of multiple parameters, the fault diagnosis module adjusts the loss weights of various parameters through the data balance module; The fault diagnosis module after adjusting the loss weight processes the equipment operating parameters to obtain the fault diagnosis results.
6. The fault diagnosis method for household appliances according to any one of claims 1 to 5, characterized in that: Before determining the data status of the equipment operating parameters, the fault diagnosis method also includes: The equipment operating parameters are preprocessed to remove noise data in the equipment operating parameters and unify the equipment operating parameters under the same dimension.
7. The fault diagnosis method for household appliances according to any one of claims 1 to 5, characterized in that: The first diagnostic model or the second diagnostic model is trained as follows: Obtain the labeled training sample set, which includes a training set, a validation set, and a test set; Based on the preset optimization algorithm, in each iteration, a batch of data is randomly selected from the training set to train the basic model to obtain the first diagnostic model or the second diagnostic model; During the training process, the performance of the basic model is continuously tested based on the sample data in the test set. If the performance of the basic model remains unchanged within the set number of iterations, the training of the basic model is stopped. During the training process, the performance of the basic model is continuously tested based on the sample data in the validation set, and the hyperparameters of the basic model are dynamically adjusted according to the performance test results.
8. A fault diagnosis device for household appliances, characterized in that: include: An acquisition module, configured to acquire device operating parameters of the home device; A determination module, configured to determine data conditions of equipment operation parameters; A determination module, configured to determine a target diagnostic model in the first diagnostic model and the second diagnostic model according to data of the device operation parameters; A processing module is configured to input the equipment operation parameters into the target diagnosis model for processing to obtain a fault diagnosis result; Among them, the first diagnostic model is a pre-built model that can process data anomalies of equipment operating parameters and perform fault diagnosis based on the processed equipment operating parameters; the second diagnostic model is a pre-built model that can perform fault diagnosis based on equipment operating parameters with normal data.
9. A fault diagnosis device for household appliances, comprising a processor and a memory storing program instructions, characterized in that: The processor is configured to execute the fault diagnosis method for a household appliance according to any one of claims 1 to 7 when running the program instructions.
10. An electronic device, characterized in that: include: Equipment body; The fault diagnosis device for household appliances as claimed in claim 8 or 9 is installed on the appliance body.