A fault analysis method and system based on a device model

Abstract

This application relates to the technical field of fault analysis, and discloses a fault analysis method and system based on an equipment model, including: loading the equipment model of the switch cabinet; obtaining the real-time operation data of the switch cabinet, and disassembling the real-time operation data into sub-data under different working forms; for any sub-data, sending the sub-data to the corresponding form sub-model; collecting the simulation data obtained by each form sub-model after running the sub-data according to the time sequence relationship, and constructing multiple groups of simulation time sequence arrays based on the collected simulation data; generating the fault analysis results of each group of simulation time sequence arrays one by one; obtaining the fault analysis result of the last simulation time sequence array, and determining the fault analysis result as the fault analysis result of the switch cabinet. The beneficial effect is more refined monitoring and fault analysis.

Description

Technical Field

[0001] This application relates to the technical field of fault analysis, and particularly to a fault analysis method and system based on an equipment model. Background Art

[0002] The current fault analysis methods for switchgear have limitations in many aspects. First, manual judgment is easily affected by human factors such as the experience and emotions of operators, resulting in inaccurate or untimely diagnosis of complex or rare faults. Second, traditional simple simulation technologies usually only consider typical working states and ignore the changes of equipment under complex working conditions, making it difficult to comprehensively reflect the combined effects of various operating states and fault modes of the equipment, thus affecting the analysis accuracy.

[0003] Therefore, the current problem that needs to be continuously solved is how to improve the accuracy and precision of fault analysis and achieve more refined monitoring and fault analysis. Summary of the Invention

[0004] This application provides a fault analysis method and system based on an equipment model, which solves the technical problem of how to improve the accuracy and precision of fault analysis and achieves the technical effect of more refined monitoring and fault analysis.

[0005] To achieve the above object, the main technical solutions adopted in this application include:

[0006] In the first aspect, an embodiment of this application provides a fault analysis method based on an equipment model, and the method includes:

[0007] Load the equipment model of the switchgear, where the equipment model includes multiple morphological sub-models of the switchgear in different working states, and each of the morphological sub-models is arranged according to the timing relationship of the working states;

[0008] Obtain the real-time operation data of the switchgear and disassemble the real-time operation data into sub-data in different working states;

[0009] For any sub-data, send the sub-data to the corresponding morphological sub-model to run the sub-data through the morphological sub-model;

[0010] According to the timing relationship, collect the simulation data obtained by each of the morphological sub-models after running the sub-data, and construct multiple groups of simulation timing arrays based on the collected simulation data;

[0011] Generate the fault analysis results of each group of simulation timing arrays one by one. Among them, when generating the fault analysis results of the current simulation timing array, obtain the historical fault analysis results of the previous simulation timing array located before the current simulation timing array, and use the historical fault analysis results as the guiding label to perform fault prediction on the current simulation timing array;

[0012] Obtain the fault analysis results of the last simulation timing array, and determine the fault analysis results as the fault analysis results of the switch cabinet.

[0013] A fault analysis method based on an equipment model provided in this embodiment provides a basis for subsequent accurate fault analysis by loading multiple morphological sub-models, which are arranged according to different working forms of the equipment and their timing relationships. Obtain real-time operation data and disassemble it into sub-data related to different working forms to ensure the pertinence and efficiency of the data. The sub-data is sent to the corresponding morphological sub-models for processing to ensure that each sub-data can perform simulation operations under appropriate working conditions, and can effectively simulate the performance of the switch cabinet under different working forms. Collect simulation data according to the timing relationship and construct multiple groups of simulation timing arrays, providing sufficient data support for subsequent fault analysis. For each group of simulation timing arrays, by combining historical fault analysis results, it is possible to predict the faults of the current simulation data with the help of fault modes in previous data. This historical data-oriented analysis method not only improves the accuracy of prediction but also enhances the robustness of the model. Finally, by analyzing the fault analysis results of the last simulation timing array, a comprehensive evaluation and judgment of the switch cabinet faults are provided.

[0014] Optionally, sending the sub-data to the corresponding morphological sub-models includes:

[0015] Identify the target working form to which the sub-data belongs, and obtain the target timing identifier corresponding to the target working form;

[0016] Transmit the sub-data to the morphological sub-model with the target timing identifier.

[0017] In this embodiment, by identifying the target working form of the sub-data and obtaining the corresponding target timing identifier, it is ensured that each sub-data is allocated to the appropriate processing path as needed. Transmitting the sub-data to the morphological sub-model with the corresponding timing identifier for further processing realizes the automation of data processing and precise timing management, not only improving the processing efficiency of the system but also enhancing flexibility, and can optimize the processing process under different working forms and time series requirements.

[0018] Optionally, constructing multiple groups of simulation timing arrays according to the collected simulation data includes:

[0019] Identify the timing identifiers covered by the multiple morphological sub-models. For any current timing identifier, obtain the simulation data corresponding to the current timing identifier, as well as the simulation data corresponding to each of the other timing identifiers before the current timing identifier, and splice the obtained simulation data according to the timing identifiers.

[0020] Fill the spliced data into a preset first field, and fill the number of copies of the spliced simulation data into a preset second field, and use the array composed of the first field and the second field as the simulation timing array corresponding to the current timing identifier.

[0021] In the process of timing identifier identification and simulation data acquisition in this embodiment, it is possible to comprehensively collect and understand the operating states of the switchgear at different time nodes, ensuring a complete grasp of the historical evolution process. By splicing the simulation data at different time nodes, a continuous time series is formed, revealing the changing trend of the operating state over time. Filling the spliced data into the preset first field and filling the number of copies of the simulation data into the second field can ensure the structuring and standardization of the data during storage, making subsequent processing more efficient. The obtained simulation timing array serves as a data carrier, facilitating subsequent tasks such as timing analysis and fault diagnosis.

[0022] Optionally, the fault analysis results of each group of simulation timing arrays are generated as follows:

[0023] For adjacent first and second simulation timing arrays, obtain the first fault analysis result of the first simulation timing array, and set the first fault analysis result as the guiding label of the second simulation timing array.

[0024] Input the second simulation timing array with the guiding label set into the trained fusion network. The fusion network includes a label configuration network and a result prediction network. Among them, the label configuration network extracts the features of the guiding label and determines a weight matrix that matches the features of the guiding label. The weight matrix is loaded into the result prediction network so that the result prediction network loaded with the weight matrix performs fault prediction on the second simulation timing array.

[0025] In this embodiment, by processing the adjacent first simulation time series array and second simulation time series array, first, the fault analysis result of the first simulation time series array is obtained and used as the guiding label for the second simulation time series array. This ensures that the analysis of the second simulation time series array can be based on the fault patterns in the first simulation time series array, thereby providing effective reference information for subsequent fault prediction. The guiding label is input into the trained fusion network. The fusion network includes a label configuration network and a result prediction network. The main function of the label configuration network is to extract features from the guiding label and determine the weight matrix that matches these features through the learning process. After training, the weight matrix can reflect the influence degree of different features in the guiding label on fault prediction. Subsequently, this weight matrix is loaded into the result prediction network to enhance the performance of the network in the fault prediction task. The embodiment of the present application can effectively associate and share information between different simulation time series arrays, improving the accuracy of the prediction result.

[0026] Optionally, the matching relationship between the guiding label and the weight matrix is determined in the following manner:

[0027] Obtain time series training samples, where the time series training samples include guiding label samples, time series array samples, and prediction labels;

[0028] Use the label configuration network to extract features from the guiding label samples, and fuse the extracted features with the time series array samples to obtain fused samples;

[0029] Use the result prediction network to predict the fused samples to obtain fault prediction results;

[0030] Calculate the error between the fault prediction result and the prediction label, and adjust the weight matrix of the result prediction network based on the error so that when the result prediction network after adjusting the weight matrix predicts the fused samples again, the obtained fault prediction result matches the prediction label;

[0031] Match the guiding label samples with the adjusted weight matrix.

[0032] In this embodiment, by obtaining time-series training samples including guiding label samples, time-series array samples, and prediction labels, it provides basic data for subsequent model training. Then, the label configuration network extracts features from the guiding label samples. The extracted features can effectively reveal the key features of the fault mode. These features are fused with the time-series array samples to form fused samples. The fused samples are input into the result prediction network for fault prediction to obtain preliminary fault prediction results. By calculating the error between the prediction result and the prediction label, the weight matrix of the result prediction network is adjusted according to this error. This process enables the model to more accurately learn the fault mode in the prediction task, thereby optimizing the prediction result. Finally, the guiding label samples are matched with the adjusted weight matrix to ensure that the information sharing and interaction in the entire prediction process reach the optimal state. It not only improves the accuracy and reliability of the fault prediction model, but also enhances the adaptability of the model, enabling it to handle various different fault modes and time-series data, and improving the intelligent level and generalization ability of fault prediction.

[0033] Optionally, fusing the extracted features with the time-series array samples to obtain fused samples includes:

[0034] Identifying each row of data in the extracted features and splicing the data of each row according to the row number to obtain a corresponding spliced vector;

[0035] Using the spliced vector as the suffix of the time-series array sample to form a fused sample.

[0036] In this embodiment, each row of data in the features is identified and extracted, and then these data are spliced according to the row number to obtain a corresponding spliced vector. It can effectively combine the row data in the features in order, retaining the time-series information and its internal structure in the time-series data, fully reflecting the order of each row of data, thereby improving the accuracy and effectiveness of subsequent analysis. Then, the spliced vector is used as the suffix of the time-series array sample and combined with the original time-series data to form a fused sample. Through this splicing method, the time-series array sample and the extracted feature information are organically fused, enabling each sample to incorporate more valuable information obtained from feature extraction on the basis of containing time-series information. This fusion enhances the model's ability to understand time-series data and improves the accuracy and robustness of fault prediction.

[0037] Optionally, the real-time operation data of the switch cabinet includes at least one of online temperature measurement data of new contacts, online detection data of switching voltage and current, online monitoring data of mechanical characteristics, and sensing data of partial discharge sensors.

[0038] In a second aspect, an embodiment of the present application provides a fault analysis system based on an equipment model, and the system includes:

[0039] A model loading unit for loading the device model of the switchgear cabinet, where the device model includes multiple morphological sub-models of the switchgear cabinet in different working states, and each of the morphological sub-models is arranged according to the chronological relationship of the working states;

[0040] A data disassembling unit for obtaining the real-time operation data of the switchgear cabinet and disassembling the real-time operation data into sub-data in different working states;

[0041] An operation unit for, for any sub-data, sending the sub-data to the corresponding morphological sub-model to run the sub-data through the morphological sub-model;

[0042] An array construction unit for collecting the simulation data obtained by each of the morphological sub-models after running the sub-data according to the chronological relationship, and constructing multiple groups of simulation chronological arrays based on the collected simulation data;

[0043] A fault analysis unit for generating the fault analysis results of each group of simulation chronological arrays one by one. Among them, when generating the fault analysis result of the current simulation chronological array, obtaining the historical fault analysis result of the previous simulation chronological array located before the current simulation chronological array, and using the historical fault analysis result as a guiding label to perform fault prediction on the current simulation chronological array;

[0044] A result determination unit for obtaining the fault analysis result of the last simulation chronological array and determining the fault analysis result as the fault analysis result of the switchgear cabinet.

[0045] Optionally, the fault analysis unit is specifically configured to, for adjacent first and second simulation chronological arrays, obtain the first fault analysis result of the first simulation chronological array and set the first fault analysis result as the guiding label of the second simulation chronological array; input the second simulation chronological array with the guiding label set into the trained fusion network, where the fusion network includes a label configuration network and a result prediction network. Among them, the label configuration network extracts the features of the guiding label and determines a weight matrix that matches the features of the guiding label, and the weight matrix is loaded into the result prediction network so that the result prediction network loaded with the weight matrix performs fault prediction on the second simulation chronological array.

[0046] Optionally, the fault analysis unit determines the matching relationship between the guiding label and the weight matrix in the following manner: obtaining a timing training sample, which includes a guiding label sample, a timing array sample, and a prediction label; using a label configuration network to extract features from the guiding label sample, and fusing the extracted features with the timing array sample to obtain a fused sample; using a result prediction network to predict the fused sample to obtain a fault prediction result; calculating the error between the fault prediction result and the prediction label, and adjusting the weight matrix of the result prediction network based on the error, so that when the result prediction network after the weight matrix adjustment predicts the fused sample again, the obtained fault prediction result matches the prediction label; matching the guiding label sample with the adjusted weight matrix. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] In order to more clearly illustrate the specific embodiments of the present application or the technical solutions in the prior art, the following will briefly introduce the drawings required for the description of the specific embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0048] Figure 1 It is a flowchart of a fault analysis method based on a device model provided by an embodiment of the present application;

[0049] Figure 2 It is a flowchart of step S5 provided by an embodiment of the present application;

[0050] Figure 3 It is a flowchart of step S7 provided by an embodiment of the present application;

[0051] Figure 4 It is a flowchart of step S9 provided by an embodiment of the present application;

[0052] Figure 5 It is a flowchart of a matching relationship determination method provided by an embodiment of the present application;

[0053] Figure 6 It is a flowchart of obtaining a fused sample provided by an embodiment of the present application;

[0054] Figure 7 It is a frame of a fault analysis system based on a device model provided by an embodiment of the present application;

[0055] Figure 8 It is a structural schematic diagram of a computer device provided by an embodiment of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0056] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the following will clearly and completely describe the technical solutions in the embodiments of this application with reference to the accompanying drawings in the embodiments of this application. Obviously, the described embodiments are part of the embodiments of this application, rather than all of them. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative efforts belong to the scope of protection of this application.

[0057] According to the embodiments of this application, an embodiment of a fault analysis method based on an equipment model is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. And although the logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in a different order than here.

[0058] The current fault analysis methods for switchgear have limitations in many aspects. First of all, manual fault analysis usually relies on the experience and intuition of operators and is easily affected by human factors such as fatigue, emotion, or knowledge blind spots. This results in the analysis results being inaccurate or untimely when facing complex or rare faults. For example, operators may not be able to cover all possible fault modes completely or may ignore some potential abnormal signals, thus affecting the accuracy and reliability of fault diagnosis.

[0059] In addition, some systems adopt fault analysis methods based on traditional simulation models. These simulation models are usually relatively simple and often only consider some typical working forms of the switchgear, ignoring the details and changes of the equipment under complex working conditions. This simple simulation method is difficult to comprehensively cover the various working forms of the equipment and cannot accurately reflect the complex behavior of the switchgear in a changing environment, thereby affecting the accuracy of fault analysis. For example, a simple simulation model may not be able to simulate the combined effects of multiple fault modes and cannot dynamically adapt to the real-time changing state of the equipment, resulting in a large deviation between the prediction result and the actual fault.

[0060] To solve the above problems, an embodiment of a fault analysis method based on an equipment model is provided in this embodiment. Figure 1 The following is a flowchart of a fault analysis method based on an equipment model provided by the embodiments of this application. As Figure 1 shown, this process includes the following steps:

[0061] Step S1: Load the equipment model of the switchgear. The equipment model includes multiple morphological sub-models of the switchgear in different working forms, and each morphological sub-model is arranged according to the time sequence relationship of the working forms.

[0062] Specifically, the working mode refers to the set of different working states of the switchgear during actual operation and operation processes. These states include the operation process of the switchgear equipment, the changes during operation, and the different working states when the circuit breaker is switched on and off. The morphological sub-model refers to the model established for each specific working mode, constructed through digital twin technology, and used to simulate and analyze the behavior of the switchgear under a specific working mode. Each morphological sub-model is arranged according to the chronological relationship of the working mode to ensure that the behavior of the entire switchgear can be continuously and accurately simulated.

[0063] Step S3: Obtain the real-time operation data of the switchgear and disassemble the real-time operation data into sub-data under different working modes.

[0064] Specifically, the operation data of the switchgear is collected in real time through sensors and monitoring systems. These data may include at least one of the online temperature measurement data of new contacts, the online detection data of switching voltage and current, the online monitoring data of mechanical characteristics, and the sensing data of partial discharge sensors. The data is disassembled according to the time stamp and state change of the real-time operation data, including slicing the data according to time periods based on the time when events occur (such as closing, opening, fault, etc.). The data within each time period is classified as a specific working mode. By monitoring the state changes of the switchgear (such as closing, opening, fault alarm, etc.), these state change points are used as the markers for data slicing. For example, after detecting a closing operation, relevant data can be extracted from that moment until the next operation or state change point. Each data slice is labeled with corresponding tags, such as "closing operation", "over-temperature state", "fault protection state", etc. This allows for quick positioning of specific working mode data during subsequent analysis. The disassembled data is stored in a database, usually organized and indexed by time and working mode. By tagging the data, it is convenient for querying and further analysis.

[0065] Step S5: For any sub-data, send the sub-data to the corresponding morphological sub-model to run the sub-data through the morphological sub-model.

[0066] Specifically, each morphological sub-model is designed for a specific working mode and can simulate the behavior of the switchgear in that mode. The sub-data is sent to the corresponding morphological sub-model, and after receiving the sub-data, the morphological sub-model performs physical simulation operation.

[0067] Step S7: According to the chronological relationship, collect the simulation data obtained by each morphological sub-model after running the sub-data, and construct multiple groups of simulation time series arrays based on the collected simulation data.

[0068] Specifically, according to the timing relationship, collect the simulation data of each morphological sub-model after collecting the operation sub-data to ensure that the simulation data is ordered in the time series. By collecting these data in chronological order, the evolution process of the switchgear during the entire operation cycle can be reflected. After collecting these time-series data, multiple simulation time-series arrays will be constructed based on the collected data. A simulation time-series array refers to a data structure in which simulation data is organized in chronological order, and each array corresponds to a set of simulation data under a specific time series.

[0069] Step S9, generate the fault analysis results of each group of simulation time-series arrays one by one. Among them, when generating the fault analysis results of the current simulation time-series array, obtain the historical fault analysis results of the previous simulation time-series array before the current simulation time-series array, and use the historical fault analysis results as the guiding labels to perform fault prediction on the current simulation time-series array.

[0070] Specifically, analyze each group of simulation time-series arrays to generate fault analysis results. When generating the fault analysis results of the current simulation time-series array, it is necessary to obtain the historical fault analysis results of the previous simulation time-series array before the current simulation time-series array. These historical results are used as guiding labels to provide context information for the current fault prediction, and can effectively utilize historical fault data to optimize the fault prediction of real-time simulation data, improving the accuracy and reliability of the prediction. This method combines historical fault data and real-time simulation data in order to obtain more accurate prediction results.

[0071] Step S11, obtain the fault analysis results of the last simulation time-series array, and determine the fault analysis results as the fault analysis results of the switchgear.

[0072] Specifically, after completing the fault analysis of all simulation time-series arrays, collect the fault analysis results obtained from the last simulation time-series array. This fault analysis result is the fault analysis result of the switchgear, and this result takes into account the influence of the latest operation data and historical fault analysis results.

[0073] A fault analysis method based on a device model provided in this embodiment provides a basis for subsequent accurate fault analysis by loading multiple morphological sub-models, which are arranged according to different working states of the device and their timing relationships. Real-time operation data is obtained and disassembled into sub-data related to different working states to ensure the pertinence and efficiency of the data. The sub-data is sent to the corresponding morphological sub-model for processing to ensure that each sub-data can perform simulation operations under appropriate working conditions, and can effectively simulate the performance of the switchgear under different working states. The simulation data is collected according to the timing relationship and multiple groups of simulation timing arrays are constructed, providing sufficient data support for subsequent fault analysis. For each group of simulation timing arrays, by combining the historical fault analysis results, the fault prediction of the current simulation data can be carried out with the help of the fault modes in the previous data. This historical data-oriented analysis method not only improves the accuracy of prediction but also enhances the robustness of the model. Finally, through the analysis of the fault analysis results of the last simulation timing array, a comprehensive evaluation and judgment of the switchgear fault are provided.

[0074] Figure 2 The flowchart of step S5 provided in the embodiment of this application may include the following steps:

[0075] Step S51, identify the target working state to which the sub-data belongs, and obtain the target timing identifier corresponding to the target working state.

[0076] Specifically, in order to identify the working state of the switchgear corresponding to each sub-data in the real-time operation data, machine learning algorithms such as decision trees, random forests, and support vector machines can be used for classification. First, key features that can represent different working states are extracted from the sub-data. For example, characteristic parameters of the closing and opening current waveforms, the vibration frequency and amplitude of the circuit breaker, etc. can be extracted. These features can reflect different working states of the switchgear. Then, the extracted features are formed into a feature vector as the input of the model. Using the method of supervised learning, the model is trained with historical data to obtain a pre-trained classification model. This model can judge the working state corresponding to the sub-data according to the input feature vector. Finally, through the trained classification model, each sub-data is classified into different working states according to the feature vector, so as to identify the target working state to which the sub-data belongs.

[0077] After the target working state is identified, the corresponding target timing identifier needs to be obtained. Here, a unique timing identifier corresponding to each working state of the switchgear needs to be pre-allocated, that is, each working state is identified. Then, according to the predefined mapping relationship of working state - timing identifier, the target timing identifier corresponding to the target working state is found.

[0078] Step S53, transmit the sub-data to the morphological sub-model with the target timing identifier.

[0079] Specifically, once the target working form of the sub-data and the corresponding target timing identifier are determined, the sub-data is transmitted to the form sub-model with the corresponding timing identifier to ensure that the data can be correctly sent to the form sub-model for further analysis and processing.

[0080] Compared with Figure 1 the embodiment shown, in this embodiment, by identifying the target working form of the sub-data and obtaining the corresponding target timing identifier, it is ensured that each sub-data is allocated to the appropriate processing path as needed. Transmitting the sub-data to the form sub-model with the corresponding timing identifier for further processing realizes the automation of data processing and precise timing management. It not only improves the processing efficiency of the system but also enhances the flexibility, and can optimize the processing process under different working forms and time series requirements.

[0081] Figure 3 The flowchart of step S7 provided by the embodiment of the present application, this process may include the following steps:

[0082] Step S71, identify the timing identifiers covered by multiple form sub-models. For any current timing identifier, obtain the simulation data corresponding to the current timing identifier, and obtain the simulation data corresponding to each of the other timing identifiers before the current timing identifier, and splice the obtained simulation data according to the timing identifiers.

[0083] Specifically, first identify the timing identifiers covered by multiple form sub-models. These timing identifiers represent different working forms. Each timing identifier corresponds to specific simulation data, and the simulation data records the operation of the switchgear at that timing node. For any current timing identifier, obtain the relevant simulation data. This simulation data usually contains key status information at this timing point, such as the operating parameters of the equipment, fault information, or other monitoring indicators, used to describe the current working state. At the same time, it is also necessary to obtain the simulation data corresponding to the other timing identifiers before the current timing identifier. By obtaining the simulation data corresponding to these historical timing identifiers, it is possible to comprehensively understand how the current state has evolved and the change trends in the historical data. Finally, splice all the obtained simulation data in the order of the timing identifiers. This process concatenates the simulation data at different time nodes to form a continuous time series, providing a comprehensive view of the operation process of the switchgear. Through the spliced simulation data, a more in-depth analysis and prediction of the entire working form can be carried out, identifying potential problems, optimizing operation strategies, and providing more accurate fault warnings and maintenance plans.

[0084] Step S73: Fill the spliced data into a preset first field, and fill the number of copies of the spliced simulation data into a preset second field. Then, use the array composed of the first field and the second field as the simulation time series array corresponding to the current time series identifier.

[0085] Specifically, fill the spliced simulation data into the preset first field, which represents all the simulation data under the current time series identifier. Fill the number of copies of the spliced simulation data into the preset second field, which records the quantity of simulation data under the current time series identifier and helps to understand the integrity and coverage of the data. Combine the first field and the second field to form the simulation time series array corresponding to the current time series identifier. This array not only contains the simulation data but also the statistical information of the data, providing comprehensive data support for fault analysis.

[0086] Compared with Figure 1 the embodiment shown, the process of time series identifier recognition and simulation data acquisition in this embodiment enables the comprehensive collection and understanding of the operating states of switchgears at different time nodes, ensuring a complete grasp of the historical evolution process. By splicing the simulation data at different time nodes, a continuous time series is formed, revealing the changing trend of the operating state over time. Filling the spliced data into the preset first field and filling the number of copies of the simulation data into the second field can ensure the structuring and standardization of the data during storage, making subsequent processing more efficient. The obtained simulation time series array serves as a data carrier, facilitating subsequent time series analysis, fault diagnosis, and other tasks.

[0087] Figure 4 The flowchart of step S9 provided by the embodiment of the present application may include the following steps:

[0088] Step S91: For adjacent first and second simulation time series arrays, obtain the first fault analysis result of the first simulation time series array and set the first fault analysis result as the guiding label of the second simulation time series array.

[0089] Specifically, the fault analysis result obtained from the first simulation time series array is actually based on historical data and previous simulation results, reflecting the existing fault patterns and rules. Use these fault analysis results as guiding labels and apply them to the second simulation time series array. This guiding label acts as a representative of historical fault data and provides guidance for subsequent simulation data. In the subsequent fault prediction process, historical fault data can not only be combined with real-time data to help identify possible faults but also improve the prediction accuracy through known fault patterns. Therefore, this method can effectively improve the prediction accuracy and ensure that fault prediction not only depends on current data but also incorporates historical experience.

[0090] Step S93: Input the second simulated timing array with the guiding label into the trained fusion network. The fusion network includes a label configuration network and a result prediction network. Among them, the label configuration network extracts features from the guiding label and determines a weight matrix that matches the features of the guiding label. The weight matrix is loaded into the result prediction network so that the result prediction network loaded with the weight matrix performs fault prediction on the second simulated timing array.

[0091] Specifically, first, the label configuration network is responsible for extracting features from the guiding label. This process can be achieved through a deep learning model. For example, a convolutional neural network (CNN) or a long short-term memory network (LSTM) can be used to identify the key features in the guiding label. These features can represent the important information of the guiding label and provide support for subsequent weight matrix determination and fault prediction. Once the features of the guiding label are extracted, the label configuration network determines a weight matrix that matches these features. This weight matrix is obtained through learning during the training process and can reflect the influence degree of the guiding label features on fault prediction. The determination of the weight matrix can be achieved through optimization algorithms such as stochastic gradient descent (SGD) or the Adam optimizer. These algorithms can adjust the weights during the training process to minimize the prediction error. Next, the determined weight matrix is loaded into the result prediction network, which can be an LSTM network or other models suitable for time series prediction. The result prediction network loaded with the weight matrix can utilize the information of the guiding label to improve the accuracy of fault prediction. The loading of the weight matrix is equivalent to providing prior knowledge to the network, enabling the network to pay more attention to the features that match the historical fault patterns when processing the second simulated timing array. The result prediction network loaded with the adapted weight matrix performs fault prediction on the second simulated timing array. This prediction process combines historical fault data and real-time simulation data in order to obtain more accurate prediction results.

[0092] It should be noted that by constructing the association relationship between the guiding label and the weight matrix, an adapted weight matrix can be determined according to the current guiding label. This process is similar to training a neural network, where the features of the guiding label are used to adjust the weights of the result prediction network to adapt to the current prediction task. The role of the weight matrix in the result prediction network is to adjust the response of the network, making it pay more attention to the features that match the historical fault patterns. In this way, the result prediction network can better capture the key factors that may lead to faults. Through this method, historical fault data can be used to optimize the fault prediction of real-time simulation data, improving the accuracy and reliability of the prediction.

[0093] And Figure 1Compared with the embodiments shown, in this embodiment, by processing adjacent first simulation time series arrays and second simulation time series arrays, the fault analysis result of the first simulation time series array is first obtained and used as the guiding label for the second simulation time series array. This ensures that the analysis of the second simulation time series array can be based on the fault patterns in the first simulation time series array, thereby providing effective reference information for subsequent fault prediction. The guiding label is input into the trained fusion network. The fusion network includes a label configuration network and a result prediction network. The main function of the label configuration network is to extract features from the guiding label and determine the weight matrix that matches these features through a learning process. After training, the weight matrix can reflect the influence degree of different features in the guiding label on fault prediction. Subsequently, this weight matrix is loaded into the result prediction network to enhance the performance of the network in the fault prediction task. The embodiment of the present application can effectively associate and share information between different simulation time series arrays, improving the accuracy of the prediction result.

[0094] Figure 5 The flowchart of the matching relationship determination method provided by the embodiment of the present application may include the following steps:

[0095] Step S931: Obtain time series training samples, which include guiding label samples, time series array samples, and prediction labels.

[0096] Specifically, the guiding label samples are extracted from historical data, which are the fault records of past switchgear and are used to indicate past fault events. The time series array samples are real-time operation data corresponding to the guiding labels, such as time series data of the voltage, current, temperature, etc. of the switchgear. The prediction labels are the expected outputs corresponding to the time series array samples, usually known fault labels, and are used to train the model to identify faults.

[0097] Step S933: Use the label configuration network to extract features from the guiding label samples, and fuse the extracted features with the time series array samples to obtain fused samples.

[0098] Use a label configuration network, such as a convolutional neural network (CNN) or a long short-term memory network (LSTM), to identify and extract features in the guiding label samples, combine the extracted features with the time series array samples to form a new data set, that is, the fused sample, which contains the original time series information and historical fault features.

[0099] Step S935: Use the result prediction network to predict the fused samples to obtain the fault prediction result.

[0100] Specifically, the fused sample is input into a result prediction network, such as a Convolutional Neural Network (CNN) or a Long Short-Term Memory Network (LSTM), and the network outputs a fault prediction result based on the fused sample.

[0101] Step S937, calculate the error between the fault prediction result and the prediction label, and adjust the weight matrix of the result prediction network based on the error, so that when the result prediction network after the weight matrix adjustment predicts the fused sample again, the obtained fault prediction result matches the prediction label.

[0102] Specifically, to calculate the error between the fault prediction result and the prediction label, Mean Squared Error (MSE) or Cross-Entropy Error can be used, and then according to the calculated error, an optimization algorithm (such as SGD or Adam) is used to adjust the weight matrix of the result prediction network until the error does not exceed a preset threshold.

[0103] Step S939, match the guiding label sample with the adjusted weight matrix.

[0104] Specifically, match the adjusted weight matrix with the guiding label sample to obtain the corresponding relationship between the weight matrix and the guiding label sample, for subsequent optimization of the network's prediction ability for new samples.

[0105] Compared with Figure 4 the embodiment shown, in this embodiment, by obtaining a time-series training sample containing a guiding label sample, a time-series array sample, and a prediction label, it provides basic data for subsequent model training. Then the label configuration network extracts features from the guiding label sample, and the extracted features can effectively reveal the key features of the fault mode. These features are fused with the time-series array sample to form a fused sample. The fused sample is input into the result prediction network for fault prediction to obtain a preliminary fault prediction result. By calculating the error between the prediction result and the prediction label, the weight matrix of the result prediction network is adjusted according to this error. This process enables the model to more accurately learn the fault mode in the prediction task, thereby optimizing the prediction result. Finally, the guiding label sample is matched with the adjusted weight matrix to ensure that the information sharing and interaction in the entire prediction process reach the optimal state. It not only improves the accuracy and reliability of the fault prediction model, but also enhances the adaptability of the model, enabling it to handle various different fault modes and time-series data, and improving the intelligent level and generalization ability of fault prediction.

[0106] Figure 6 The flowchart for obtaining the fused sample provided by the embodiment of the present application may include the following steps:

[0107] Step S9331, identify the data in each row of the extracted features, and splice the data in each row according to the row number to obtain the corresponding spliced vector.

[0108] Specifically, each row of data is identified from the features extracted from the label configuration network. The identified feature row data are concatenated in the order of their original row numbers. This concatenation process integrates multi-dimensional feature information into a one-dimensional vector so as to be able to fuse with the time series array samples. After the concatenation is completed, a concatenated vector is obtained, which contains the continuous information of all row data and provides a basis for subsequent fusion.

[0109] Step S9333, using the concatenated vector as the suffix of the time series array sample to form a fusion sample.

[0110] Specifically, using the concatenated vector as the suffix of the time series array sample to form a fusion sample. This means appending the concatenated vector to the end of the time series array sample to form a longer vector, which contains both time series data and feature information extracted from the guiding labels. The fusion sample integrates the original time series data and the feature information of the guiding labels and provides more comprehensive input data for the result prediction network.

[0111] Compared with Figure 5 the embodiment shown, in this embodiment, each row of data in the features is identified and extracted, and then these data are concatenated according to the row numbers to obtain the corresponding concatenated vector. It can effectively combine the row data in the features in order, retain the time series information and its internal structure in the time series data, fully reflect the order of each row of data, and thus improve the accuracy and effectiveness of subsequent analysis. Then, the concatenated vector is used as the suffix of the time series array sample and combined with the original time series data to form a fusion sample. Through this concatenation method, the time series array sample and the extracted feature information are organically fused, so that each sample can add more valuable information obtained from feature extraction on the basis of containing time series information. This fusion enhances the model's ability to understand time series data and improves the accuracy and robustness of fault prediction.

[0112] Correspondingly, please refer to Figure 7 the frame of a fault analysis system based on a device model provided by an embodiment of the present application. The system includes:

[0113] A model loading unit 101, configured to load a device model of a switch cabinet. The device model includes a plurality of form sub-models in different working forms of the switch cabinet, and each form sub-model is arranged according to the time series relationship of the working forms;

[0114] A data disassembling unit 103, configured to obtain real-time operation data of the switch cabinet and disassemble the real-time operation data into sub-data in different working forms;

[0115] The operation unit 105 is configured to send the sub-data to the corresponding morphological sub-model for any sub-data, so as to run the sub-data through the morphological sub-model;

[0116] The array construction unit 107 is configured to collect the simulation data obtained by each morphological sub-model after running the sub-data according to the time sequence relationship, and construct multiple groups of simulation time sequence arrays according to the collected simulation data;

[0117] The fault analysis unit 109 is configured to generate the fault analysis results of each group of simulation time sequence arrays one by one. When generating the fault analysis result of the current simulation time sequence array, obtain the historical fault analysis result of the previous simulation time sequence array located before the current simulation time sequence array, and use the historical fault analysis result as the guiding label to perform fault prediction on the current simulation time sequence array;

[0118] The result determination unit 111 is configured to obtain the fault analysis result of the last simulation time sequence array and determine the fault analysis result as the fault analysis result of the switch cabinet.

[0119] In some alternative embodiments, the fault analysis unit 109 is specifically configured to, for adjacent first and second simulation time sequence arrays, obtain the first fault analysis result of the first simulation time sequence array and set the first fault analysis result as the guiding label of the second simulation time sequence array; input the second simulation time sequence array with the set guiding label into the trained fusion network, where the fusion network includes a label configuration network and a result prediction network. The label configuration network extracts features from the guiding label and determines a weight matrix that matches the features of the guiding label. The weight matrix is loaded into the result prediction network so that the result prediction network loaded with the weight matrix performs fault prediction on the second simulation time sequence array.

[0120] In some alternative embodiments, the fault analysis unit 109 determines the matching relationship between the guiding label and the weight matrix in the following manner: obtain a time sequence training sample, where the time sequence training sample includes a guiding label sample, a time sequence array sample, and a prediction label; use the label configuration network to extract features from the guiding label sample and fuse the extracted features with the time sequence array sample to obtain a fused sample; use the result prediction network to predict the fused sample to obtain a fault prediction result; calculate the error between the fault prediction result and the prediction label, and adjust the weight matrix of the result prediction network based on the error so that when the result prediction network after adjusting the weight matrix predicts the fused sample again, the obtained fault prediction result matches the prediction label; match the guiding label sample with the adjusted weight matrix.

[0121] In some alternative embodiments, sending the sub-data to the corresponding morphological sub-model includes:

[0122] Identify the target working form to which the sub-data belongs, and obtain the target timing identifier corresponding to the target working form;

[0123] Transmit the sub-data to the form sub-model with the target timing identifier.

[0124] In some alternative embodiments, according to the collected simulation data, construct multiple groups of simulation timing arrays, including:

[0125] Identify the timing identifiers covered by multiple form sub-models. For any current timing identifier, obtain the simulation data corresponding to the current timing identifier, and obtain the simulation data corresponding to each of the other timing identifiers before the current timing identifier, and splice the obtained simulation data according to the timing identifiers;

[0126] Fill the spliced data into a preset first field, and fill the number of copies of the spliced simulation data into a preset second field, and use the array composed of the first field and the second field as the simulation timing array corresponding to the current timing identifier.

[0127] In some alternative embodiments, the fault analysis results of each group of simulation timing arrays are generated one by one, including:

[0128] For adjacent first and second simulation timing arrays, obtain the first fault analysis result of the first simulation timing array, and set the first fault analysis result as the guiding label of the second simulation timing array;

[0129] Input the second simulation timing array with the guiding label set into the trained fusion network. The fusion network includes a label configuration network and a result prediction network. Among them, the label configuration network extracts features from the guiding label and determines a weight matrix that matches the features of the guiding label. The weight matrix is loaded into the result prediction network so that the result prediction network loaded with the weight matrix performs fault prediction on the second simulation timing array.

[0130] In some alternative embodiments, the real-time operation data of the switch cabinet includes at least one of online temperature measurement data of a new type of contact, online detection data of switching-on and switching-off voltage and current, online monitoring data of mechanical characteristics, and sensing data of a partial discharge sensor.

[0131] The further function descriptions of the above-mentioned various modules and units are the same as those in the corresponding above-mentioned embodiments, and will not be elaborated here.

[0132] A fault analysis system based on a device model in this embodiment is presented in the form of a functional unit. Here, the unit refers to an ASIC (Application Specific Integrated Circuit) circuit, a processor and a memory that execute one or more software or fixed programs, and / or other devices that can provide the above functions.

[0133] Please refer to Figure 8 , Figure 8 which is a schematic structural diagram of a computer device provided by an embodiment of the present application. As Figure 8 shown, the computer device includes: one or more processors 10, a memory 20, and an interface for connecting each component, including a high-speed interface and a low-speed interface. Each component communicates with each other using different buses and can be installed on a common motherboard or installed in other ways as needed. The processor can process instructions executed within the computer device, including instructions stored in the memory or on the memory to display graphical information of the GUI on an external input / output device (such as a display device coupled to the interface). In some alternative embodiments, if necessary, multiple processors and / or multiple buses can be used together with multiple memories and multiple memories. Similarly, multiple computer devices can be connected, and each device provides some necessary operations (for example, as a server array, a set of blade servers, or a multi-processor system). Figure 8 In

[0134] which, one processor 10 is taken as an example.

[0135] The memory 20 stores instructions executable by at least one processor 10, so that the at least one processor 10 executes the method shown in the above embodiment.

[0136] The memory 20 may include a program storage area and a data storage area. Among them, the program storage area may store an operating system and application programs required for at least one function; the data storage area may store data created according to the use of the computer device and the like. In addition, the memory 20 may include high-speed random access memory, and may also include non-transitory memory, such as at least one magnetic disk storage device, a flash memory device, or other non-transitory solid-state storage devices. In some alternative embodiments, the memory 20 may optionally include a memory remotely provided with respect to the processor 10, and these remote memories may be connected to the computer device through a network. Examples of the above-mentioned network include but are not limited to the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0137] The memory 20 may include volatile memory, such as random access memory; the memory may also include non-volatile memory, such as flash memory, a hard disk, or a solid-state drive; the memory 20 may further include a combination of the above types of memory.

[0138] The computer device further includes a communication interface 30 for the computer device to communicate with other devices or a communication network.

[0139] The embodiments of the present application further provide a computer-readable storage medium. The methods according to the embodiments of the present application can be implemented in hardware, firmware, or be implemented as computer code that can be recorded on a storage medium, or be implemented as computer code originally stored in a remote storage medium or a non-transitory machine-readable storage medium and downloaded through a network and will be stored in a local storage medium, so that the methods described herein can be stored in such software processes on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. Among them, the storage medium may be a magnetic disk, an optical disk, a read-only memory, a random access memory, a flash memory, a hard disk, or a solid-state drive, etc.; further, the storage medium may also include a combination of the above types of memory. It can be understood that a computer, a processor, a microprocessor controller, or programmable hardware includes a storage component that can store or receive software or computer code, and when the software or computer code is accessed and executed by the computer, the processor, or the hardware, the methods shown in the above embodiments are implemented.

[0140] The systems or units illustrated in the above embodiments may be specifically implemented by a computer chip or an entity, or by a product having a certain function. A typical implementation device is a computer. Specifically, the computer may be, for example, a personal computer, a laptop computer, a cellular phone, a camera phone, a smart phone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or any combination of these devices.

[0141] For the convenience of description, when describing the above device, various units are described separately according to their functions. Of course, when implementing the present application, the functions of each unit can be implemented in one or more software and / or hardware.

[0142] Those skilled in the art should understand that the embodiments of the present application can be provided as methods and systems. Therefore, the present application can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0143] The present application is described with reference to the flowcharts and / or block diagrams of methods and systems according to the embodiments of the present application. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate a device for implementing the specified functions in Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0144] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory generate a manufactured article including an instruction device, and the instruction device implements the specified functions in Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0145] These computer program instructions can also be loaded onto a computer or other programmable data processing device, so that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process, and thus the instructions executed on the computer or other programmable device provide steps for implementing the specified functions in Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0146] It should also be noted that the term "comprise", "include" or any other variant thereof is intended to cover non-exclusive inclusion, such that a process, method, commodity or device comprising a series of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, commodity or device. Without further limitation, an element defined by the phrase "comprising an..." does not exclude the presence of additional identical elements in the process, method, commodity or device comprising said element.

[0147] Each embodiment in this specification is described in a progressive manner. For the identical or similar parts among the embodiments, reference can be made to each other, and the differences between each embodiment and other embodiments are emphasized. In particular, for the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and reference can be made to the relevant part of the method embodiment for the related content.

[0148] The above description is only for the embodiments of the present application and is not intended to limit the present application. For those skilled in the art, various modifications and changes can be made to the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included within the scope of the claims of the present application.

[0149] Although the embodiments of the present application are described in conjunction with the accompanying drawings, those skilled in the art can make various modifications and variations without departing from the spirit and scope of the present application, and such modifications and variations fall within the scope defined by the appended claims.

Claims

1. A fault analysis method based on a device model, characterized in that: The method comprises: Loading a device model of a switch cabinet, wherein the device model includes a plurality of state sub-models of the switch cabinet in different working states, and each of the state sub-models is arranged according to a time sequence relationship of the working state; Acquire the real-time operation data of the switch cabinet, and decompose the real-time operation data into sub-data under different working modes; For any sub-data, the sub-data is sent to the corresponding morphological sub-model to run the sub-data through the morphological sub-model; According to the time series relationship, the simulation data obtained after the sub-data of each morphological sub-model is run are collected, and multiple groups of simulation time series arrays are constructed according to the collected simulation data; Generate fault analysis results for each group of simulation timing arrays one by one, wherein when generating the fault analysis results of the current simulation timing array, obtain the historical fault analysis results of the previous simulation timing array before the current simulation timing array, and use the historical fault analysis results as guide labels to perform fault prediction on the current simulation timing array; A fault analysis result of the last simulation time series array is obtained, and the fault analysis result is determined as the fault analysis result of the switch cabinet.

2. The method according to claim 1, characterized in that Sending the sub-data to the corresponding morphological sub-model includes: Identify the target working form to which the sub-data belongs, and obtain the target timing identifier corresponding to the target working form; The sub-data is transmitted to the morphological sub-model having the target timing identification.

3. The method according to claim 1 or 2, characterized in that: According to the collected simulation data, constructing multiple sets of simulation timing arrays includes: Identify the timing identifiers covered by the multiple morphological sub-models, and for any current timing identifier, obtain the simulation data corresponding to the current timing identifier, and obtain the simulation data corresponding to each of the other timing identifiers before the current timing identifier, and splice the obtained simulation data according to the timing identifiers; The spliced ​​data is filled into a preset first field, and the number of spliced ​​simulation data is filled into a preset second field, and the array formed by the first field and the second field is used as the simulation timing array corresponding to the current timing identifier.

4. The method according to claim 1, characterized in that The fault analysis results generated for each group of simulation timing arrays include: For a first simulation timing array and a second simulation timing array that are adjacent to each other, obtaining a first fault analysis result of the first simulation timing array, and setting the first fault analysis result as a guide label of the second simulation timing array; The second simulation timing array with the guide label set is input into the trained fusion network, wherein the fusion network includes a label configuration network and a result prediction network, wherein the label configuration network extracts features of the guide label and determines a weight matrix that matches the features of the guide label, and the weight matrix is ​​loaded into the result prediction network, so that the result prediction network loaded with the weight matrix performs fault prediction on the second simulation timing array.

5. The method according to claim 4, characterized in that The matching relationship between the guide label and the weight matrix is ​​determined by: Acquire a time series training sample, wherein the time series training sample includes a guide label sample, a time series array sample, and a prediction label; Extracting features from the guide label samples using a label configuration network, and fusing the extracted features with the time series array samples to obtain fused samples; Using a result prediction network to predict the fusion sample, to obtain a fault prediction result; Calculating the error between the fault prediction result and the prediction label, and adjusting the weight matrix of the result prediction network based on the error, so that when the result prediction network after the weight matrix adjustment predicts the fusion sample again, the obtained fault prediction result matches the prediction label; The guide label samples are matched with the adjusted weight matrix.

6. The method according to claim 5, characterized in that The extracted features are fused with the time series array samples to obtain fused samples including: Identify each row of data in the extracted features, and concatenate each row of data according to the row number to obtain a corresponding concatenation vector; The concatenated vector is used as a suffix of the time series array sample to form a fused sample.

7. The method according to claim 1, characterized in that The real-time operation data of the switch cabinet includes at least one of online temperature measurement data of new contacts, online detection data of opening and closing voltage and current, online monitoring data of mechanical characteristics, and sensing data of partial discharge sensors.

8. A fault analysis system based on a device model, characterized in that: The system comprises: A model loading unit, used to load a device model of the switch cabinet, wherein the device model includes a plurality of form sub-models of the switch cabinet in different working modes, and each of the form sub-models is arranged according to a time sequence relationship of the working modes; A data decomposition unit, used for acquiring the real-time operation data of the switch cabinet, and decomposing the real-time operation data into sub-data under different working modes; An operating unit, for sending any sub-data to a corresponding morphological sub-model, so as to operate the sub-data through the morphological sub-model; An array construction unit, used to collect the simulation data obtained after running the sub-data of each morphological sub-model according to the time series relationship, and construct multiple groups of simulation time series arrays according to the collected simulation data; A fault analysis unit, used to generate fault analysis results of each group of simulation timing arrays one by one, wherein when generating the fault analysis results of the current simulation timing array, a historical fault analysis result of a previous simulation timing array before the current simulation timing array is obtained, and the historical fault analysis result is used as a guide label to perform fault prediction on the current simulation timing array; The result determination unit is used to obtain the fault analysis result of the last simulation time series array, and determine the fault analysis result as the fault analysis result of the switch cabinet.

9. The system according to claim 8, characterized in that The fault analysis unit is specifically used to obtain a first fault analysis result of the first simulation timing array and a second adjacent simulation timing array, and set the first fault analysis result as a guide label of the second simulation timing array; input the second simulation timing array with the guide label set into the trained fusion network, the fusion network includes a label configuration network and a result prediction network, wherein the label configuration network extracts features of the guide label and determines a weight matrix that matches the features of the guide label, and the weight matrix is ​​loaded into the result prediction network, so that the result prediction network loaded with the weight matrix performs fault prediction on the second simulation timing array.

10. The system according to claim 9, characterized in that The fault analysis unit determines the matching relationship between the guide label and the weight matrix in the following manner: obtaining a time series training sample, wherein the time series training sample includes a guide label sample, a time series array sample, and a prediction label; using a label configuration network to extract features from the guide label sample, and fusing the extracted features with the time series array sample to obtain a fused sample; using a result prediction network to predict the fused sample to obtain a fault prediction result; Calculate the error between the fault prediction result and the prediction label, and adjust the weight matrix of the result prediction network based on the error, so that when the result prediction network after the weight matrix adjustment predicts the fusion sample again, the obtained fault prediction result matches the prediction label; match the guide label sample with the adjusted weight matrix.

Images