Mining method, device and equipment of power equipment failure data and storage medium

Abstract

The application provides a power equipment fault data mining method, device and equipment and a storage medium. The method comprises: performing classification processing on power signal data to be mined to obtain target power signal data corresponding to a target data type from the power signal data; determining a target data mining model matched with the target data type based on a preset model database, wherein the target data mining model comprises at least one data mining sub-model stored in the model database; and performing fault data mining processing on the target power signal data by using the target data mining model to obtain power equipment fault data. The method can effectively reduce the calculation amount in the mining process, improve the power equipment fault data mining precision and efficiency, and improve the mining effect by screening out target power signal data to be mined and using a corresponding and adaptive target data mining model for data mining on the target power signal data.

Description

Technical Field

[0001] This application relates to the field of data processing technology, and in particular to a method, apparatus, equipment and storage medium for mining power equipment fault data. Background Technology

[0002] Power equipment failures can lead to unplanned outages, causing economic losses for power companies. Current power equipment failure detection technologies involve extensive computation during data mining and export, resulting in complex and low-tolerance-for-error processes that are prone to errors and failures. Furthermore, broken data links during the mining process cannot be reassembled, leading to data loss and poor mining results. Summary of the Invention

[0003] In view of this, embodiments of this application provide a method, apparatus, device, and storage medium for mining power equipment fault data, which can reduce the amount of computation, effectively improve the accuracy and efficiency of power equipment fault data mining, reduce data loss, and improve mining results.

[0004] A first aspect of this application provides a method for mining power equipment fault data, comprising: classifying power signal data to be mined; obtaining target power signal data corresponding to a target data type from the power signal data; determining a target data mining model matching the target data type based on a preset model database, wherein the target data mining model includes at least one data mining sub-model stored in the preset model database; and performing fault data mining processing on the target power signal data using the target data mining model to obtain power equipment fault data, wherein the target power signal data contains one or more power equipment fault data.

[0005] In conjunction with the first aspect, in the first possible implementation of the first aspect, before the step of determining the target data mining model matching the target data type based on the preset model database, the method further includes: collecting power equipment fault data samples based on preset data types to obtain several sample sets, wherein one sample set corresponds to one data type; for each data type, performing feature deep learning using power equipment fault data in the sample set corresponding to the data type to obtain fault features corresponding to the data type; for each data type, calculating the data volume storage of the fault features corresponding to the data type to obtain data volume storage information corresponding to the fault features; for each data type, performing memory allocation processing based on the data volume storage information corresponding to the fault features to obtain a first allocated memory corresponding to the data type; for each data type, building a neural network model based on the fault features and the first allocated memory and training the neural network model to obtain a data mining sub-model matching the data type; mapping and associating the data mining sub-model matching the data type with the data type and storing it in the preset model database.

[0006] In a second possible implementation of the first aspect, in conjunction with the first possible implementation of the first aspect, the step of building a neural network model based on the fault features corresponding to the data type and the first allocated memory, and training the neural network model to obtain a data mining sub-model corresponding to the data type includes: performing secondary classification processing on the fault features corresponding to the data type so that the data type contains several secondary categories; setting weight parameters for the several secondary categories to obtain several first neural network weight values, wherein the several first neural network weight values ​​correspond one-to-one with the several secondary categories; performing memory allocation processing on the several secondary categories based on the several first neural network weight values ​​and the first allocated memory corresponding to the data type to obtain second allocated memory corresponding to each of the several secondary categories; building a neural network model based on the several secondary categories and the first neural network weight values ​​and second allocated memory corresponding to each of the several secondary categories; and training the neural network model using power equipment fault data in the sample set corresponding to the data type to obtain a data mining sub-model corresponding to the data type.

[0007] In conjunction with the first aspect, in a third possible implementation of the first aspect, the step of determining a target data mining model matching the target data type based on the preset model database includes: searching the preset model database according to the target data type to obtain at least one data mining sub-model that has a mapping relationship with the target data type; receiving model selection information from a user and determining a target model structure according to the user's model selection information, wherein the target model structure is any one of a feedback network model structure, a feedforward network model structure, or a self-organizing network model structure; and importing the obtained at least one data mining sub-model into the target model structure for model building to obtain a target data mining model matching the target data type.

[0008] In conjunction with the third possible implementation of the first aspect, in the fourth possible implementation of the first aspect, the step of importing the obtained at least one data mining sub-model into the target model structure for model building to determine a target data mining model matching the target data type further includes: when obtaining multiple data mining sub-models, setting weights for the multiple data mining sub-models so that the multiple data mining models have their own corresponding second neural network weight values; and importing the multiple data mining models into the target model structure for model building in a weighted combination manner according to the second neural network weight values ​​corresponding to each of the multiple data mining models to obtain a target data mining model matching the target data type.

[0009] In a fifth possible implementation of the first aspect, in conjunction with the first aspect or the first, second, third, or fourth possible implementations of the first aspect, after the step of using the target data mining model to perform fault data mining processing on the target power signal data to obtain power equipment fault data, wherein the target power signal data contains one or more power equipment fault data, the method further includes: performing maximum byte data extraction processing on the one or more power equipment fault data and performing numbering processing based on the maximum byte data to obtain the number corresponding to each of the one or more power equipment fault data; using the number corresponding to each of the one or more power equipment fault data as the node name, establishing a decision tree node in a preset decision tree to store the one or more power equipment fault data in the preset decision tree.

[0010] In conjunction with the first aspect or the first, second, third, or fourth possible implementations of the first aspect, in the sixth possible implementation of the first aspect, after the step of using the target data mining model to perform fault data mining processing on the target power signal data to obtain power equipment fault data, wherein the target power signal data contains one or more power equipment fault data, the method further includes: performing data preprocessing on the one or more power equipment fault data respectively, wherein the data preprocessing includes data attribute generalization processing and data attribute reduction processing, corresponding to obtaining one or more first power equipment fault data, wherein the first power equipment fault data is power equipment fault data after data preprocessing; recombining and recording the one or more first power equipment fault data to obtain a second power fault data, and storing the second power fault data as the mining data corresponding to the power signal data in a preset power equipment fault database.

[0011] A second aspect of this application provides a device for mining power equipment fault data, comprising: a classification module for classifying power signal data to be mined and obtaining target power signal data corresponding to a target data type from the power signal data; a matching module for determining a target data mining model that matches the target data type based on a preset model database, wherein the target data mining model includes at least one data mining sub-model stored in the preset model database; and a mining module for performing fault data mining processing on the target power signal data using the target data mining model to obtain power equipment fault data, wherein the target power signal data contains one or more power equipment fault data.

[0012] A third aspect of this application provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the electronic device, wherein the processor executes the computer program to implement the steps of the method for mining power equipment fault data provided in the first aspect.

[0013] A fourth aspect of this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the power equipment fault data mining method provided in the first aspect.

[0014] The present application provides a method, apparatus, electronic device, and storage medium for mining power equipment fault data, which has the following beneficial effects:

[0015] This application classifies the power signal data to be mined, extracting target power signal data corresponding to the target data type. Then, based on a pre-set model database, a target data mining model matching the target data type is determined. This target data mining model includes at least one data mining sub-model stored in the pre-set model database. The target data mining model is then used to perform fault data mining on the power signal data to obtain power equipment fault data. Based on this method, using corresponding matching data mining models for different types of power signal data effectively reduces the computational load of the mining process, improves the accuracy and efficiency of power equipment fault data mining, and enhances the mining effect. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of this application, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 A flowchart illustrating the implementation of a method for mining power equipment fault data, provided in an embodiment of this application;

[0018] Figure 2 A flowchart illustrating a method for constructing a model database in the method for mining power equipment fault data provided in this application embodiment;

[0019] Figure 3 A flowchart illustrating a method for obtaining a data mining sub-model from power equipment fault data mining methods provided in this application embodiment;

[0020] Figure 4 A flowchart illustrating a method for determining a target data mining model in the power equipment fault data mining method provided in this application embodiment;

[0021] Figure 5 A schematic diagram of another method for determining the target data mining model in the method for mining power equipment fault data provided in the embodiments of this application;

[0022] Figure 6 A flowchart illustrating a method for storing power equipment fault data in the power equipment fault data mining method provided in this application embodiment;

[0023] Figure 7 A schematic flowchart illustrating another method for storing power equipment fault data in the power equipment fault data mining method provided in this application embodiment;

[0024] Figure 8 A basic structural block diagram of a power equipment fault data mining device provided in this application embodiment;

[0025] Figure 9 This is a basic structural block diagram of an electronic device provided in an embodiment of this application. Detailed Implementation

[0026] 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.

[0027] Power equipment generates a large amount of data during operation. To collect fault data generated during this process, it is necessary to mine and process this data, requiring a data processing platform for identifying power equipment faults. The power equipment fault data mining method provided in this application is mainly applied to a data processing platform for identifying power equipment faults, aiming to effectively reduce the computational load of the mining process, improve the accuracy and efficiency of power equipment fault data mining, and enhance the mining effect.

[0028] Please see Figure 1 , Figure 1 This application provides a flowchart of a method for mining power equipment fault data. Specifically, it may include the following steps S11 to S13.

[0029] S11: Classify the power signal data to be mined, and obtain the target power signal data corresponding to the target data type from the power signal data.

[0030] In this embodiment, the power signal data refers to the signal data generated by power equipment during operation. The data processing platform can acquire the power signal data to be mined through one or more data acquisition methods, including but not limited to data download and wireless data transmission. It is understood that the power signal data to be mined can be pre-collected delayed data or real-time data collected in real time. In this embodiment, multiple different data types can be pre-set and corresponding classification rules can be configured. After acquiring the power signal data to be mined, the pre-configured classification rules are used to classify the power signal data, grouping power signal data of the same data type together, i.e., one data type corresponds to one group of power signal data. Thus, users can specify the target data type according to their mining needs and acquire the group of power signal data corresponding to the target data type as the target power signal data. For example, in this embodiment, the pre-set data types can include, but are not limited to, continuous signal data types and discrete signal data types, energy signal data types and power signal data types, time-domain signal data types and frequency-domain signal data types, etc., and multiple data types can also be set according to signal strength. Taking multiple data types set according to signal strength as an example, multiple data types can be set according to signal strength. Each data type corresponds to a different signal strength range. The classification rule is to compare the signal strength of the power signal data with the corresponding signal strength range of each data type, and determine whether the signal strength of the power signal data falls within the signal strength range corresponding to a certain data type. Power signal data whose signal strength falls within the signal strength range corresponding to a certain data type are assigned to the group corresponding to that data type.

[0031] S12: Based on the preset model database, determine a target data mining model that matches the target data type, wherein the target data mining model includes at least one data mining sub-model stored in the preset model database.

[0032] In this embodiment, a preset model database stores various data mining sub-models applicable to different data types, and a mapping relationship is established between data types and data mining sub-models. After obtaining the target data type corresponding to the power signal data by classifying the power signal data, the preset model database can be queried. Based on the mapping relationship between data types and data mining sub-models in the model database, one or more data mining sub-models matching the target data type can be obtained from the preset model database to construct the target data mining model. If multiple data mining sub-models matching the target data type are obtained from the preset model database, different weights can be set for these multiple data mining sub-models, and they can be weighted and combined according to their weights to determine the target data mining model. This target data mining model is used to extract data features and identify fault types from the power signal data of the target data type, and to generate power equipment fault data based on the data features and fault types. In this embodiment, the data mining sub-models stored in the preset model database are neural network models obtained by deep learning training on a large number of power signal data samples of the target data type until convergence.

[0033] S13: The target power signal data is processed by the target data mining model to obtain power equipment fault data, wherein the target power signal data contains one or more power equipment fault data.

[0034] In this embodiment, after determining the target data mining model, the target power signal data is input into the target data mining model. The target data mining model uses its built-in feature extraction method to extract features from the target power signal data, obtaining data features of the target power signal data. These data features are then compared with fault features corresponding to several different fault types included in the target data mining model. It is determined whether the similarity between the data feature and the fault feature reaches a preset threshold. If the similarity between the data feature and one or more fault features reaches the preset threshold, a signal data segment containing the data feature is extracted from the power signal data. Based on the fault type corresponding to one or more fault features whose similarity to the data feature reaches the preset threshold, a fault type label is set for the signal data segment. This signal data segment with the fault type label is identified as power equipment fault data in the target power signal data and output by the target data mining model. It should be noted that in this embodiment, when multiple data features are extracted from the target power signal data, multiple power equipment fault data can be obtained.

[0035] As can be seen from the above, the power equipment fault data mining method provided in this application first classifies the power signal data to obtain the target power signal data corresponding to the target data type. Then, based on a preset model database, one or more data mining sub-models matching the target data type are selected from the model database and combined to determine the target data mining model matching the target data type. Finally, the target data mining model is used to perform fault data mining processing on the target power signal data to obtain the power equipment fault data. Based on this method, by classifying and filtering the target power signal data to be mined, and using a corresponding and suitable target data mining model for data mining, the computational load of the mining process can be effectively reduced, the accuracy and efficiency of power equipment fault data mining can be improved, and the mining effect can be enhanced.

[0036] In some embodiments of this application, please refer to Figure 2 , Figure 2 This is a flowchart illustrating a method for constructing a model database in the power equipment fault data mining method provided in this application embodiment. Specifically, constructing the model database may include the following steps S21 to S26.

[0037] S21: Collect power equipment fault data samples based on preset data types to obtain several sample sets, where one data type corresponds to one sample set;

[0038] S22: For each data type, feature deep learning is performed using power equipment fault data in the sample set corresponding to the data type to obtain the fault features corresponding to the data type;

[0039] S23: For each data type, perform data volume storage calculation on the fault characteristics corresponding to the data type to obtain data volume storage information corresponding to the fault characteristics;

[0040] S24: For each data type, memory allocation is performed based on the data volume occupancy information corresponding to the fault characteristics to obtain the first allocated memory corresponding to the data type;

[0041] S25: For each data type, build a neural network model based on the fault characteristics corresponding to the data type and the first allocated memory to obtain a data mining sub-model matching the data type;

[0042] S26: Map and associate the data mining sub-model corresponding to the data type with the data type and store it in the preset model database.

[0043] In this embodiment, the preset data types may include, but are not limited to, continuous signal data types and discrete signal data types, energy signal data types and power signal data types, time-domain signal data types and frequency-domain signal data types, and various data types set according to signal strength. In this embodiment, the data processing platform can set up a data acquisition interface to collect data according to preset data types, collecting power equipment fault data samples and obtaining different sample sets for different data types. One data type corresponds to one sample set. After collecting the power equipment fault data samples, for each data type, feature deep learning can be performed using the power equipment fault data in the corresponding sample set to obtain the fault features corresponding to the data type. Then, the data volume storage usage of the fault features corresponding to the data type is calculated to obtain the data volume storage information corresponding to the fault features. Memory allocation processing is then performed based on the data volume storage information corresponding to the fault features to obtain the first allocated memory corresponding to the data type. Finally, a neural network model is built based on the fault features corresponding to the data type and the first allocated memory, and the neural network model is trained to obtain the matching data mining sub-model corresponding to the data type. It is understood that for each data type, a corresponding matching data mining sub-model can be obtained by building a neural network model, and one data type can correspond to one or more data mining sub-models. After obtaining the data mining sub-model, the data mining sub-model can be mapped and associated with data types and stored in a preset model database for use in subsequent data mining processes.

[0044] In some embodiments of this application, please refer to Figure 3 , Figure 3 This is a flowchart illustrating a method for obtaining a data mining sub-model using a power equipment fault data mining method provided in this application embodiment. Specifically, obtaining the data mining sub-model may include the following steps S31 to S34.

[0045] S31: Based on the fault characteristics corresponding to the data type, perform secondary classification processing on the fault characteristics so that the data type contains several secondary categories;

[0046] S32: Set weight parameters for the plurality of secondary categories to obtain a plurality of first neural network weight values, wherein the plurality of first neural network weight values ​​correspond one-to-one with the plurality of secondary categories;

[0047] S33: Based on the weight values ​​of the plurality of first neural networks and the first allocated memory corresponding to the data type, perform memory allocation processing on the plurality of secondary categories to obtain the second allocated memory corresponding to each of the plurality of secondary categories;

[0048] S34: Based on the several secondary categories and the first neural network weight values ​​and second allocated memory corresponding to each of the several secondary categories, a neural network model is built. The neural network model is trained using power equipment fault data in the sample set corresponding to the data type to obtain the data mining sub-model matching the data type.

[0049] In this embodiment, when building a neural network model for each data type, the fault features can be further refined into secondary classifications based on the fault characteristics corresponding to the data type, so that the data type contains several secondary categories. Then, weight parameters are set for each of these secondary categories to obtain several first neural network weight values, each corresponding one-to-one with a secondary category. In this embodiment, the first neural network weight values ​​can be entered by the user into the data processing platform. After obtaining the first neural network weight values ​​corresponding to each secondary category, memory allocation is performed on the secondary categories based on these first neural network weight values ​​and the first allocated memory corresponding to the data type, obtaining a second allocated memory for each secondary category. Furthermore, a neural network model is built based on these secondary categories, their corresponding first neural network weight values, and the second allocated memory. The neural network model is trained using power equipment fault data from the sample set corresponding to the data type, thereby obtaining a data mining sub-model matching the data type. In this embodiment, neural network models can be trained separately for different secondary categories, so that each secondary category corresponds to a data mining sub-model, resulting in multiple corresponding data mining sub-models for each data type. Alternatively, a neural network model containing several sub-networks can be built based on these secondary categories. By training this neural network model containing several sub-networks, a data mining sub-model corresponding to a data type can be obtained.

[0050] In some embodiments of this application, please refer to Figure 4 , Figure 4 This is a flowchart illustrating a method for determining a target data mining model in the power equipment fault data mining method provided in this application embodiment. Specifically, determining the target data mining model may include the following steps S41 to S43.

[0051] S41: Based on the target data type, search the preset model database to obtain at least one data mining sub-model that has a mapping relationship with the target data type;

[0052] S42: Receive the user's model selection information, and determine the target model structure based on the user's model selection information, wherein the target model structure is any one of a feedback network model structure, a feedforward network model structure, or a self-organizing network model structure;

[0053] S43: Import the obtained at least one data mining sub-model into the target model structure to build the model and obtain a target data mining model that matches the target data type.

[0054] In this embodiment, the target data type can be any one or more of the preset data types in the data processing platform, which can be determined by the user through data type selection within the platform. After determining the target data type, a preset model database can be searched based on this target data type to obtain at least one data mining sub-model that has a mapping relationship with the target data type. Furthermore, by having the user select a model in the data processing platform, the platform receives the user's model selection information and determines the target model structure based on this information. The target model structure can be any one of a feedback network model structure, a feedforward network model structure, or a self-organizing network model structure. The user can determine the model structure used to build the target data mining model through model selection. After determining the target model structure, by importing at least one obtained data mining sub-model into the determined target model structure for model building, a target data mining model matching the target data type can be obtained.

[0055] In some embodiments of this application, please refer to Figure 5 , Figure 5 This is a schematic flowchart illustrating another method for determining the target data mining model in the power equipment fault data mining method provided in this application embodiment. Specifically, determining the target data mining model may further include the following steps S51 and S52.

[0056] S51: When obtaining multiple data mining sub-models, weights are set for the multiple data mining sub-models so that the multiple data mining models have their own corresponding second neural network weight values.

[0057] S52: Based on the weight values ​​of the second neural network corresponding to each of the multiple data mining models, the multiple data mining models are imported into the target model structure in a weighted combination manner to build the model and obtain a target data mining model that matches the target data type.

[0058] In this embodiment, when determining the target data mining model, if multiple data mining sub-models have been previously obtained from a preset model database based on the target data type, the user can be instructed to input the corresponding second neural network weight values ​​of each of these sub-models into the data processing platform. This sets the weights for the multiple data mining sub-models, ensuring each model has its own corresponding second neural network weight value. Then, based on the corresponding second neural network weight values, the multiple data mining models are imported into the target model structure using a weighted combination method to build the model, obtaining a combined model. This combined model is then determined as the target data mining model that matches the target data type.

[0059] In some embodiments of this application, please refer to Figure 6 , Figure 6 This is a flowchart illustrating a method for storing power equipment fault data in the power equipment fault data mining method provided in this application embodiment. Specifically, storing power equipment fault data may include the following steps S61 and S62.

[0060] S61: Extract the maximum byte data from the one or more power equipment fault data and perform numbering based on the maximum byte data to obtain the number corresponding to each of the one or more power equipment fault data.

[0061] S62: Using the number corresponding to each of the one or more power equipment fault data as the node name, establish a decision tree node in a preset decision tree to store the one or more power equipment fault data in the preset decision tree.

[0062] In this embodiment, data mining processing using target power signal data can yield one or more power equipment fault data points. For these obtained fault data points, the maximum byte size can be extracted, and a unique identifier can be assigned based on this maximum byte size. Each fault data point corresponds to a unique identifier. After obtaining the identifiers, these identifiers are used as node names to create decision tree nodes within a pre-defined decision tree, thus storing the fault data points within the pre-defined decision tree.

[0063] In some embodiments of this application, please refer to Figure 7 , Figure 7This is a schematic flowchart illustrating another method for storing power equipment fault data in the power equipment fault data mining method provided in this application embodiment. Specifically, storing power equipment fault data may further include the following steps S71 and S72.

[0064] S71: Perform data preprocessing on the one or more power equipment fault data respectively, wherein the data preprocessing includes data attribute generalization processing and data attribute reduction processing, and obtain one or more first power equipment fault data, wherein the first power equipment fault data is the power equipment fault data after data preprocessing.

[0065] S72: Recombine and record the one or more first power equipment fault data to obtain a second power fault data, and store the second power fault data as the mining data corresponding to the power signal data in a preset power equipment fault database.

[0066] In this embodiment, when storing power equipment fault data, data preprocessing can be performed on the one or more power equipment fault data sets, such as data attribute generalization and data attribute reduction. After data preprocessing, one or more first power equipment fault data sets can be obtained, that is, the first power equipment fault data sets are the power equipment fault data sets after data preprocessing. In this embodiment, data attribute generalization specifically involves abstracting data from a lower conceptual level to a higher conceptual level based on data attributes to summarize the data. Data attribute reduction specifically involves deleting redundant or ineffective attributes while keeping the data classification information unchanged, which can be achieved using the RoughSets algorithm. After obtaining the first power equipment fault data sets, the one or more first power equipment fault data sets are further recombine and recorded to obtain a second power equipment fault data set. The second power equipment fault data set is stored as the mining data corresponding to the power signal data in a preset power equipment fault database. In this embodiment, when there is only one first power equipment fault data set, it is directly recorded as the second power equipment fault data set. When there are multiple first-level power equipment fault data, these multiple first-level power equipment fault data are recombined to form a single power fault data, and then this recombined power fault data is recorded as the second power fault data. Through this embodiment, some broken data can be recombined, reducing data loss during data mining and improving the overall application effect of the data processing platform.

[0067] In this embodiment, the quality of the mined data can be improved by performing data noise reduction and noisy data removal on the first power equipment fault data before recombining and recording the first power equipment fault data, and / or by performing data filtering on the obtained second power equipment fault data after recombining and recording the first power equipment fault data.

[0068] In some embodiments of this application, by applying the power equipment fault data mining method provided in this application primarily to a data processing platform for identifying power equipment faults, it is also possible to provide feedback on the mined data in the data processing platform, use the mined data as power equipment fault data samples to construct a model database, and optimize and update the data mining sub-models stored in the model database.

[0069] It is understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0070] In some embodiments of this application, please refer to Figure 8 , Figure 8 This is a basic structural block diagram of a power equipment fault data mining device provided in this application embodiment. In this embodiment, the device includes units used to perform the steps in the above method embodiments. Please refer to the relevant descriptions in the above method embodiments for details. For ease of explanation, only the parts relevant to this embodiment are shown. Figure 8 As shown, the power equipment fault data mining device includes a classification module 81, a matching module 82, and a mining module 83. Specifically: the classification module 81 classifies the power signal data to be mined, obtaining target power signal data corresponding to the target data type from the power signal data; the matching module 82 determines a target data mining model matching the target data type based on the preset model database, wherein the target data mining model includes at least one data mining sub-model stored in the preset model database; and the mining module 83 uses the target data mining model to perform fault data mining processing on the target power signal data to obtain power equipment fault data, wherein the target power signal data contains one or more power equipment fault data.

[0071] It should be understood that the aforementioned power equipment fault data mining device corresponds one-to-one with the aforementioned power equipment fault data mining method, and will not be elaborated here.

[0072] In some embodiments of this application, please refer to Figure 9 , Figure 9This is a basic structural block diagram of an electronic device provided in an embodiment of this application. Figure 9 As shown, the electronic device 9 of this embodiment includes a processor 91, a memory 92, and a computer program 93 stored in the memory 92 and executable on the processor 91, such as a program for a method of mining power equipment fault data. When the processor 91 executes the computer program 93, it implements the steps in each embodiment of the power equipment fault data mining method described above. Alternatively, when the processor 91 executes the computer program 93, it implements the functions of each module in the embodiment corresponding to the power equipment fault data mining device described above. Please refer to the relevant descriptions in the embodiments for details, which will not be repeated here.

[0073] For example, the computer program 93 can be divided into one or more modules (units), which are stored in the memory 92 and executed by the processor 91 to complete this application. The one or more modules can be a series of computer program instruction segments capable of performing specific functions, which describe the execution process of the computer program 93 in the electronic device 9. For example, the computer program 93 can be divided into a classification module, a matching module, and a mining module, each with its specific functions as described above.

[0074] The electronic device may include, but is not limited to, a processor 91 and a memory 92. Those skilled in the art will understand that... Figure 9 This is merely an example of electronic device 9 and does not constitute a limitation on electronic device 9. It may include more or fewer components than shown, or combine certain components, or different components. For example, the electronic device may also include input / output devices, network access devices, buses, etc.

[0075] The processor 91 can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor.

[0076] The memory 92 can be an internal storage unit of the electronic device 9, such as a hard disk or memory. The memory 92 can also be an external storage device of the electronic device 9, such as a plug-in hard disk, Smart Media Card (SMC), Secure Digital (SD) card, or Flash Card. Furthermore, the memory 92 can include both internal and external storage units of the electronic device 9. The memory 92 is used to store the computer program and other programs and data required by the electronic device. The memory 92 can also be used to temporarily store data that has been output or will be output.

[0077] It should be noted that the information interaction and execution process between the above-mentioned devices / units are based on the same concept as the method embodiments of this application. For details on their specific functions and technical effects, please refer to the method embodiments section, and they will not be repeated here.

[0078] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps described in the various method embodiments above. In this embodiment, the computer-readable storage medium can be either non-volatile or volatile.

[0079] This application provides a computer program product that, when run on a mobile terminal, enables the mobile terminal to implement the steps described in the above-described method embodiments.

[0080] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is merely an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit. Furthermore, the specific names of the functional units and modules are only for easy differentiation and are not intended to limit the scope of protection of this application. The specific working process of the units and modules in the above system can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0081] If the integrated module / unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording media, USB flash drives, portable hard drives, magnetic disks, optical disks, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc. It should be noted that the content included in the computer-readable medium can be appropriately added or removed according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media do not include electrical carrier signals and telecommunication signals.

[0082] In the above embodiments, the descriptions of each embodiment have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0083] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A method for mining fault data of power equipment, characterized in that, include: The power signal data to be mined is classified and processed, and the target power signal data corresponding to the target data type is obtained from the power signal data; Based on a preset model database, a target data mining model that matches the target data type is determined, wherein the target data mining model includes at least one data mining sub-model stored in the preset model database. The target data mining model is used to perform fault data mining on the target power signal data to obtain power equipment fault data, wherein the target power signal data contains one or more power equipment fault data. Before the step of determining the target data mining model that matches the target data type based on the preset model database, the method further includes: Based on preset data types, power equipment fault data samples are collected to obtain several sample sets, where one data type corresponds to one sample set. For each of the data types, feature deep learning is performed using power equipment fault data from the sample set corresponding to the data type to obtain the fault features corresponding to the data type. For each data type, the data volume occupancy is calculated for the fault characteristics corresponding to the data type to obtain the data volume occupancy information corresponding to the fault characteristics; For each of the data types, memory allocation is performed based on the data volume occupancy information corresponding to the fault characteristics to obtain the first allocated memory corresponding to the data type; For each data type, a neural network model is built based on the fault characteristics corresponding to the data type and the first allocated memory, and the neural network model is trained to obtain a data mining sub-model matching the data type. The data mining sub-models corresponding to the data type are mapped and associated with the data type and stored in the preset model database; One data type corresponds to multiple data mining sub-models; The step of building a neural network model based on the fault characteristics corresponding to the data type and the first allocated memory, and training the neural network model to obtain a matching data mining sub-model corresponding to the data type includes: Based on the fault characteristics corresponding to the data type, the fault characteristics are subjected to secondary classification processing so that the data type contains several secondary categories; Weight parameters are set for the plurality of secondary categories to obtain a plurality of first neural network weight values, wherein the plurality of first neural network weight values ​​correspond one-to-one with the plurality of secondary categories; Based on the weight values ​​of the plurality of first neural networks and the first allocated memory corresponding to the data type, memory allocation processing is performed on the plurality of secondary categories to obtain the second allocated memory corresponding to each of the plurality of secondary categories; A neural network model is built based on the several secondary categories and the first neural network weight values ​​and second allocated memory corresponding to each of the several secondary categories. The neural network model is trained using power equipment fault data in the sample set corresponding to the data type to obtain the data mining sub-model matching the data type. After the step of using the target data mining model to perform fault data mining processing on the target power signal data to obtain power equipment fault data, wherein the target power signal data contains one or more power equipment fault data, the method further includes: The maximum byte data of the one or more power equipment fault data is extracted and numbered based on the maximum byte data to obtain the number corresponding to each of the one or more power equipment fault data. The number corresponding to each of the one or more power equipment fault data is used as the node name, and a decision tree node is established in the preset decision tree to store the one or more power equipment fault data in the preset decision tree. After the step of using the target data mining model to perform fault data mining processing on the target power signal data to obtain power equipment fault data, wherein the target power signal data contains one or more power equipment fault data, the method further includes: Data preprocessing is performed on the one or more power equipment fault data respectively, wherein the data preprocessing includes data attribute generalization processing and data attribute reduction processing, corresponding to obtaining one or more first power equipment fault data, wherein the first power equipment fault data is the power equipment fault data after data preprocessing; The one or more first power equipment fault data are recombined and recorded to obtain a second power fault data, and the second power fault data is stored in a preset power equipment fault database as the mining data corresponding to the power signal data.

2. The method for mining power equipment fault data according to claim 1, characterized in that, The step of determining a target data mining model that matches the target data type based on the preset model database includes: Based on the target data type, search the preset model database to obtain at least one data mining sub-model that has a mapping relationship with the target data type; Receive the user's model selection information, and determine the target model structure based on the user's model selection information, wherein the target model structure is any one of a feedback network model structure, a feedforward network model structure, or a self-organizing network model structure; The obtained at least one data mining sub-model is imported into the target model structure to build the model, thereby obtaining a target data mining model that matches the target data type.

3. The method for mining power equipment fault data according to claim 2, characterized in that, The step of importing the obtained at least one data mining sub-model into the target model structure for model building to determine the target data mining model that matches the target data type further includes: When multiple data mining sub-models are obtained, weights are set for the multiple data mining sub-models so that the multiple data mining models have their own corresponding second neural network weight values. Based on the weight values ​​of the second neural network corresponding to each of the multiple data mining models, the multiple data mining models are imported into the target model structure in a weighted combination manner to build the model, thereby obtaining a target data mining model that matches the target data type.

4. A device for mining fault data of power equipment, characterized in that, include: The classification module is used to classify the power signal data to be mined and obtain the target power signal data corresponding to the target data type from the power signal data; A matching module is used to determine a target data mining model that matches the target data type based on a preset model database, wherein the target data mining model includes at least one data mining sub-model stored in the preset model database. The mining module is used to perform fault data mining processing on the target power signal data using the target data mining model to obtain power equipment fault data, wherein the target power signal data contains one or more power equipment fault data. Before the step of determining the target data mining model that matches the target data type based on the preset model database, the method further includes: Based on preset data types, power equipment fault data samples are collected to obtain several sample sets, where one data type corresponds to one sample set. For each of the data types, feature deep learning is performed using power equipment fault data from the sample set corresponding to the data type to obtain the fault features corresponding to the data type. For each data type, the data volume occupancy is calculated for the fault characteristics corresponding to the data type to obtain the data volume occupancy information corresponding to the fault characteristics; For each of the data types, memory allocation is performed based on the data volume occupancy information corresponding to the fault characteristics to obtain the first allocated memory corresponding to the data type; For each data type, a neural network model is built based on the fault characteristics corresponding to the data type and the first allocated memory, and the neural network model is trained to obtain a data mining sub-model matching the data type. The data mining sub-models corresponding to the data type are mapped and associated with the data type and stored in the preset model database; One data type corresponds to multiple data mining sub-models; The step of building a neural network model based on the fault characteristics corresponding to the data type and the first allocated memory, and training the neural network model to obtain a matching data mining sub-model corresponding to the data type includes: Based on the fault characteristics corresponding to the data type, the fault characteristics are subjected to secondary classification processing so that the data type contains several secondary categories; Weight parameters are set for the plurality of secondary categories to obtain a plurality of first neural network weight values, wherein the plurality of first neural network weight values ​​correspond one-to-one with the plurality of secondary categories; Based on the weight values ​​of the plurality of first neural networks and the first allocated memory corresponding to the data type, memory allocation processing is performed on the plurality of secondary categories to obtain the second allocated memory corresponding to each of the plurality of secondary categories; A neural network model is built based on the several secondary categories and the first neural network weight values ​​and second allocated memory corresponding to each of the several secondary categories. The neural network model is trained using power equipment fault data in the sample set corresponding to the data type to obtain the data mining sub-model matching the data type. After the step of using the target data mining model to perform fault data mining processing on the target power signal data to obtain power equipment fault data, wherein the target power signal data contains one or more power equipment fault data, the method further includes: The maximum byte data of the one or more power equipment fault data is extracted and numbered based on the maximum byte data to obtain the number corresponding to each of the one or more power equipment fault data. The number corresponding to each of the one or more power equipment fault data is used as the node name, and a decision tree node is established in the preset decision tree to store the one or more power equipment fault data in the preset decision tree. After the step of using the target data mining model to perform fault data mining processing on the target power signal data to obtain power equipment fault data, wherein the target power signal data contains one or more power equipment fault data, the method further includes: Data preprocessing is performed on the one or more power equipment fault data respectively, wherein the data preprocessing includes data attribute generalization processing and data attribute reduction processing, corresponding to obtaining one or more first power equipment fault data, wherein the first power equipment fault data is the power equipment fault data after data preprocessing; The one or more first power equipment fault data are recombined and recorded to obtain a second power fault data, and the second power fault data is stored in a preset power equipment fault database as the mining data corresponding to the power signal data.

5. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method as described in any one of claims 1 to 3.

6. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method as described in any one of claims 1 to 3.

Images