Electrical secondary equipment fault detection method, device, equipment, medium and program

Abstract

The invention provides an electrical secondary equipment fault detection method, device, equipment, medium and program, and the method comprises the steps: collecting the operation state information of electrical secondary equipment in a power grid, and obtaining the collected data; training a big data model by using the collected data to obtain a trained model; performing fault detection on the electrical secondary equipment in the current power grid according to the trained model to obtain a detection result; and determining a fault processing scheme according to a detection result, and carrying out maintenance order dispatching. Factor analysis and a big data technology are combined, a fault detection model which is dynamically adaptive, high in anti-interference capacity, accurate and efficient is constructed, complete-period and real-time fault detection of secondary equipment is achieved, limitation of a traditional detection method is effectively overcome, potential correlation among equipment operation parameters is mined in a data driving mode, and the fault detection accuracy is improved. Early fault signals are accurately captured, parameter drift of the equipment after long-term operation is adapted, and reliable guarantee is provided for safe and stable operation of the power grid secondary equipment.

Description

Technical Field

[0001] This invention relates to the field of electrical testing, and in particular to a method, apparatus, equipment, medium, and procedure for detecting faults in secondary electrical equipment. Background Technology

[0002] As the "nerve center" and "sensing terminal" of the power grid system, secondary electrical equipment undertakes core functions such as monitoring the power grid's operational status, transmitting control commands, and responding quickly to faults. Its operational stability directly determines the reliability of power supply, power quality, and operational safety of the power grid. Once a fault occurs, it may lead to problems such as protection malfunctions / failures, data transmission interruptions, and abnormal equipment shutdowns. These can range from minor local power outages to systemic power grid accidents, causing significant economic losses and social impacts. With the deepening of smart grid construction, the power grid scale continues to expand, and equipment types become increasingly diverse. The operating conditions of secondary equipment are becoming more complex, and its fault modes are exhibiting characteristics such as concealment, diversity, and transmissibility, placing higher demands on the timeliness and accuracy of fault detection.

[0003] Traditional fault detection for secondary electrical equipment mainly relies on manual inspections, periodic calibrations, and threshold alarms, which have significant limitations in practical applications. Relying on human experience and highly subjective: Traditional testing relies heavily on the professional experience of maintenance personnel to judge the status of equipment through on-site observation and instrument measurement. It is difficult to quantify and assess the operating trend of the equipment, and it is easily affected by differences in personal experience, resulting in a high rate of missed fault diagnosis and misdiagnosis. Weak anti-interference capability and difficulty in coping with complex working conditions: There are a lot of electromagnetic interference, message packet loss, parameter fluctuations and other interference factors in the secondary circuit. Traditional detection methods lack effective interference filtering mechanisms, which can easily misjudge interference signals as fault signals, or mask the true fault characteristics due to interference, thus affecting the accuracy of detection. Delayed detection, unable to capture early hidden dangers: Traditional detection is mostly periodic inspection or retrospective after failure, which makes it difficult to monitor the equipment operating status in real time and to detect early performance degradation and potential hidden dangers in a timely manner. Often, the problem is only dealt with after the fault becomes apparent, missing the best maintenance opportunity. Insufficient adaptability and difficulty in dealing with equipment parameter drift: After long-term operation, secondary equipment may experience parameter drift due to factors such as component aging and environmental changes. The thresholds and judgment standards of traditional detection methods are relatively fixed and cannot dynamically adapt to changes in equipment operating status, resulting in a significant decrease in detection accuracy after long-term operation.

[0004] Therefore, there is an urgent need for a method, device, equipment, medium, and procedure for detecting faults in secondary electrical equipment to improve the above-mentioned problems. Summary of the Invention

[0005] The purpose of this invention is to provide a method, apparatus, equipment, medium, and program for fault detection of electrical secondary equipment, which can detect faults in electrical secondary equipment throughout its entire lifecycle and in real time.

[0006] In a first aspect, the present invention provides a method for fault detection of electrical secondary equipment, comprising: Collect operating status information of electrical secondary equipment in the power grid to obtain the collected data; The collected data is used to train the big data model, resulting in a trained model. The fault detection is performed on the electrical secondary equipment in the current power grid based on the trained model, and the detection results are obtained. Based on the test results, a fault handling plan is determined and a repair order is dispatched.

[0007] Optional, also includes: The collected data undergoes preprocessing and dimensionality reduction. Preprocessing includes outlier handling, missing value handling, and standardization. Outlier handling uses the 3σ criterion to remove extreme biased data, and missing value handling uses interpolation to complete the data. Dimensionality reduction includes calculating the correlation coefficient matrix of the standardized data to determine data fit; performing eigenvalue decomposition on the correlation coefficient matrix to obtain the eigenvalues ​​of common factors; selecting effective common factors based on eigenvalues ​​greater than 1; performing orthogonal rotation on the common factors to make their meaning clearer; constructing factor vectors based on the selected common factors; and projecting the dataset onto the common factors for dimensionality reduction.

[0008] Optionally, the big data model is a big data-driven gradient boosting tree model. This model is built on the XGBoost framework and includes a feature selection layer, a gradient boosting layer, and an output layer. The feature selection layer uses a variance thresholding method to filter invalid features. The gradient boosting layer consists of three boosting units. The first boosting unit contains 64 decision tree nodes and uses L1 regularization to suppress overfitting. The second boosting unit contains 32 decision tree nodes and uses L2 regularization to optimize weights. The third boosting unit contains 16 decision tree nodes and uses cross-validation to optimize parameters. The outputs of each boosting unit are summed using a weighted fusion method.

[0009] Optionally, the output layer of the big data model is a multi-class output unit, which uses the cross-entropy loss function to quantify the difference between the predicted and actual values, and uses a parallel gradient descent optimizer to adjust the model parameters to meet the needs of efficient training with large amounts of data. An early stopping strategy is used during training to prevent overfitting.

[0010] Optionally, the collected data can be used to train the big data model to obtain a trained model, including: dividing the collected data into a sample set and a test set, and training the big data model to obtain an intermediate prediction model for each type of electrical secondary equipment; based on the intermediate prediction model for each target equipment, incremental updates and parameter optimizations are performed in combination with real-time collected equipment operation data to obtain a prediction model adapted to different secondary equipment operating conditions.

[0011] Optionally, the electrical secondary equipment includes relays, transformers, measurement and control devices, protection devices, and fault recorders; and / or the acquired data includes secondary circuit voltage, secondary circuit current, switch status, message signals, operating sequence, insulation resistance, and signal amplitude.

[0012] Secondly, the present invention provides an electrical secondary equipment fault detection device, which includes modules / units for performing any of the possible design methods described in the first aspect above. These modules / units can be implemented in hardware or by hardware executing corresponding software.

[0013] Thirdly, the present invention provides an electronic device including a memory and a processor, wherein the memory stores a program that can run on the processor, and when the program is executed by the processor, the electronic device implements a method for performing any of the possible designs described above.

[0014] Fourthly, the present invention provides a readable storage medium storing a program, which, when executed, implements a method of any possible design of any of the above aspects.

[0015] Fifthly, the present invention provides a computer program product, including a computer program that, when executed by a processor, implements the above-described method.

[0016] The beneficial effects of this invention are as follows: It collects operating status information of electrical secondary equipment in the power grid, obtaining the collected data; it uses the collected data to train a big data model, obtaining a trained model; it performs fault detection on the electrical secondary equipment in the current power grid based on the trained model, obtaining the detection results; and it determines a fault handling plan and dispatches maintenance orders based on the detection results. By combining factor analysis with big data technology, a dynamically adaptable, highly anti-interference, accurate, and efficient fault detection model is constructed, enabling real-time fault detection of secondary equipment throughout its entire lifecycle. This effectively overcomes the limitations of traditional detection methods, mines the potential correlations between equipment operating parameters through a data-driven approach, accurately captures early fault signals, and adapts to parameter drift after long-term equipment operation, providing a reliable guarantee for the safe and stable operation of power grid secondary equipment. Attached Figure Description

[0017] Figure 1This is a flowchart illustrating a method for detecting faults in secondary electrical equipment according to an embodiment of the present invention. Figure 2 This is a schematic diagram of the structure of an electrical secondary equipment fault detection device provided in an embodiment of the present invention; Figure 3 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions in the embodiments of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without inventive effort are within the scope of protection of this invention. Unless otherwise defined, the technical or scientific terms used herein should have the ordinary meaning understood by those skilled in the art. The terms "comprising" and similar expressions used herein mean that the element or object preceding the word covers the element or object listed following the word and its equivalents, but do not exclude other elements or objects.

[0019] The technical solutions of the embodiments of the present invention will be described below with reference to the accompanying drawings. In the description of the embodiments of the present invention, the terminology used in the following embodiments is for the purpose of describing specific embodiments only and is not intended to limit the present invention. The singular expressions “a,” “the,” “the,” and “this” are intended to also include expressions such as “one or more,” unless the context clearly indicates otherwise. It should also be understood that in the following embodiments of the present invention, “at least one” and “one or more” refer to one or more (including two). The term “and / or” is used to describe the relationship between related objects, indicating that three relationships can exist; for example, A and / or B can represent: A alone, A and B simultaneously, or B alone, where A and B can be singular or plural. The character “ / ” generally indicates that the preceding and following related objects are in an “or” relationship.

[0020] References to "one embodiment" or "some embodiments" in this specification mean that a particular feature, structure, or characteristic described in connection with that embodiment is included in one or more embodiments of the invention. Therefore, the phrases "in one embodiment," "in some embodiments," "in other embodiments," and "in still other embodiments" appearing in different parts of this specification do not necessarily refer to the same embodiment, but rather mean "one or more, but not all, embodiments," unless otherwise specifically emphasized. The terms "comprising," "including," "having," and variations thereof mean "including, but not limited to," unless otherwise specifically emphasized. The term "connection" includes both direct and indirect connections, unless otherwise stated. "First" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated.

[0021] In embodiments of the present invention, "exemplarily" or "for example" are used to indicate that they are examples, illustrations, or descriptions. Any embodiment or design described as "exemplarily" or "for example" in embodiments of the present invention should not be construed as being more preferred or advantageous than other embodiments or design solutions. Rather, the use of "exemplarily" or "for example" is intended to present the relevant concepts in a specific manner.

[0022] like Figure 1 As shown, the present invention provides a method for fault detection of electrical secondary equipment, comprising: S101 collects the operating status information of electrical secondary equipment in the power grid and obtains the collected data.

[0023] In some embodiments, the method further includes: preprocessing and dimensionality reduction of the collected data; preprocessing includes outlier handling, missing value handling, and standardization, wherein outlier handling uses the 3σ criterion to remove extreme bias data, and missing value handling uses interpolation to complete the data; dimensionality reduction includes calculating the correlation coefficient matrix of the standardized data and determining data suitability; performing eigenvalue decomposition on the correlation coefficient matrix to obtain the eigenvalues ​​of the common factors; screening effective common factors based on the principle that the eigenvalues ​​are greater than 1; performing orthogonal rotation on the common factors to make the meaning of the factors clearer; constructing factor vectors based on the screened common factors; and projecting the dataset onto the common factors for dimensionality reduction.

[0024] In another embodiment, the electrical secondary equipment includes relays, transformers, measurement and control devices, protection devices, and fault recorders.

[0025] In some other embodiments, the acquired data includes secondary circuit voltage, secondary circuit current, switch status, message signal, action timing, insulation resistance, and signal amplitude.

[0026] S102, use the collected data to train the big data model to obtain the trained model.

[0027] In some embodiments, training a big data model using the collected data to obtain a trained model includes: dividing the collected data into a sample set and a test set, and training the big data model to obtain an intermediate prediction model for each type of electrical secondary equipment; based on the intermediate prediction model for each target equipment, performing incremental updates and parameter optimization in conjunction with real-time collected equipment operation data to obtain a prediction model adapted to different operating conditions of secondary equipment.

[0028] In other embodiments, the big data model is a big data-driven gradient boosting tree model. This model is built on the XGBoost framework and includes a feature selection layer, a gradient boosting layer, and an output layer. The feature selection layer uses a variance thresholding method to filter invalid features. The gradient boosting layer consists of three boosting units. The first boosting unit contains 64 decision tree nodes and uses L1 regularization to suppress overfitting. The second boosting unit contains 32 decision tree nodes and uses L2 regularization to optimize weights. The third boosting unit contains 16 decision tree nodes and uses cross-validation to optimize parameters. The outputs of each boosting unit are summed using a weighted fusion method.

[0029] In some specific embodiments, the output layer of the big data model is a multi-classification output unit, which uses the cross-entropy loss function to quantify the difference between the predicted and actual values, and uses a parallel gradient descent optimizer to adjust the model parameters to meet the needs of efficient training with large amounts of data. During the training process, an early stopping strategy is used to prevent overfitting.

[0030] S103, based on the trained model, perform fault detection on the electrical secondary equipment in the current power grid and obtain the detection results.

[0031] S104. Determine the fault handling plan based on the test results and dispatch a repair order.

[0032] The advantages of this invention lie in its use of factor analysis combined with big data technology. A dataset is constructed by collecting parameters from different electrical secondary equipment in the power grid. Factor analysis is used to reduce the dimensionality of the data, eliminating interference signals in the secondary circuits and extracting core operating characteristics. Then, a big data model is trained based on the reduced-dimensional data, and finally, the trained prediction model is used to achieve accurate fault detection of the current electrical secondary equipment. The advantages are mainly reflected in three aspects: First, it adapts to the dynamic operating conditions of electrical secondary equipment and can cope with complex scenarios such as secondary circuit signal fluctuations and message interference; second, it has strong anti-interference capabilities, as factor analysis can effectively filter non-core interference parameters, and the big data model can uncover potential correlation features between parameters; third, the data is reliable and can accurately capture early fault signals, and the incremental model update mechanism can adapt to parameter drift after long-term equipment operation, achieving full-cycle detection from early hidden dangers to manifest faults.

[0033] To facilitate understanding, this embodiment further elaborates on the specific implementation process of the above method in conjunction with a specific application scenario, which includes the following steps: Step 1: Data Acquisition and Preprocessing, Dimensionality Reduction The electrical secondary equipment to be tested includes relays, transformers, measurement and control devices, protection devices, and fault recorders, comprehensively covering the core equipment types of the power grid secondary system. Through sensors and communication acquisition modules installed on the equipment body and secondary circuits, real-time equipment operating status information is collected, specifically including secondary circuit voltage, secondary circuit current, switching status (including opening and closing status, alarm contact status, etc.), message signals (including the length, transmission delay, and bit error rate of GOOSE and MMS messages), action sequence (including equipment action trigger time and action duration), insulation resistance, and signal amplitude (including peak value and RMS value of analog signals).

[0034] After data collection, the data is preprocessed as follows: extreme deviation data are removed using the 3σ criterion; by calculating the mean and standard deviation of each parameter, data exceeding the mean plus or minus three times the standard deviation are identified as outliers and removed; for missing values, interpolation is used to complete the data; for three or more consecutive missing data segments, polynomial interpolation is used to optimize and supplement the missing data based on the historical operating patterns of the equipment; finally, all preprocessed data are standardized to convert the data to a fixed interval to eliminate the influence of differences in the units of different parameters.

[0035] After preprocessing, dimensionality reduction is performed. The specific steps are as follows: 1) Calculate the correlation coefficient matrix of the 7 types of data after standardization, and use the KMO test (KMO value not less than 0.75) to determine whether the data is suitable for factor analysis; 2) Perform eigenvalue decomposition on the correlation coefficient matrix to obtain the eigenvalues ​​of multiple common factors; 3) Screen effective common factors according to the principle that the eigenvalue is greater than 1. The cumulative variance contribution rate of the screened common factors reaches 82.3%, which can fully reflect the core information of the original data; 4) Use the variance maximum orthogonal rotation method to orthogonally rotate the effective common factors, so that the original variable loading coefficients corresponding to each common factor are more concentrated and the meaning of the factors is clarified (corresponding to "electrical quantity stability factor", "communication transmission factor", "action response factor" and "insulation state factor" respectively); 5) Construct factor vectors based on the screened effective common factors, and project the original 7-dimensional dataset onto the factor vectors to obtain 4-dimensional dimensionality reduction data, realizing interference signal removal and core feature extraction.

[0036] Step 2, Big Data Model Training The big data model adopts a gradient boosting tree model built on the XGBoost framework. The model structure includes a feature selection layer, a gradient boosting layer, and an output layer. The feature selection layer uses a variance thresholding method (threshold set to 0.01) to filter invalid features, eliminating features with excessively small variance or no significant impact on equipment fault status. The gradient boosting layer consists of three boosting units: the first unit contains 64 decision tree nodes, using L1 regularization to suppress overfitting; the second unit contains 32 decision tree nodes, using L2 regularization to optimize weights; and the third unit contains 16 decision tree nodes, using 5-fold cross-validation to optimize parameters. The outputs of each boosting unit are aggregated using a weighted fusion method, with weights dynamically allocated based on the accuracy of each unit on the test set (0.4, 0.35, and 0.25, respectively).

[0037] The model training process is as follows: The preprocessed and dimensionality-reduced 4D data is divided into a sample set and a test set in a 7:3 ratio. The sample set contains 100,000 normal operation data points and 50,000 fault data points (covering 20 common fault types, including relay contact adhesion, transformer polarity error, communication interruption of measurement and control devices, setting drift of protection devices, and abnormal sampling of fault recorders). The model is trained using the sample set to obtain 5 intermediate prediction models for relays, transformers, measurement and control devices, protection devices, and fault recorders. In practical applications, based on the intermediate prediction model of each target device, incremental updates and parameter optimizations are performed using real-time collected equipment operation data. A model update is triggered every 10,000 new data points. The online gradient descent algorithm is used to adjust the model parameters to ensure that the model adapts to the dynamic changes in equipment operating conditions. Finally, 5 dedicated prediction models adapted to different secondary equipment operating conditions are obtained.

[0038] The model's output layer is a multi-class output unit. It uses the cross-entropy loss function to quantify the difference between the predicted and actual values, and uses a parallel gradient descent optimizer to adjust the model parameters. It supports 10-thread parallel training and is suitable for efficient training of large amounts of data. During training, an early stopping strategy is adopted. When the accuracy of the validation set does not improve for 5 consecutive epochs, training is stopped to prevent the model from overfitting.

[0039] Step 3, Fault Detection and Repair Dispatch Five trained specialized prediction models are deployed to the power grid secondary equipment operation and maintenance monitoring platform. These models receive real-time operational status data (after preprocessing and dimensionality reduction) of relays, transformers, monitoring and control devices, protection devices, and fault recorders within the power grid. This data is then input into the corresponding models for fault detection. The models calculate the matching degree between the input data and various fault characteristics, outputting the fault type (e.g., "relay contact adhesion," "transformer polarity error"), fault confidence level (ranging from 0 to 1, with a confidence level not lower than 0.8 considered a valid fault), and fault severity level (divided into Level I - emergency fault, Level II - important fault, and Level III - general fault), thus forming a complete detection result.

[0040] The operation and maintenance monitoring platform automatically determines fault handling solutions based on the detection results: For Level I emergency faults (such as protection device setting drift or communication gateway interruption), an emergency shutdown command is immediately triggered, and a maintenance dispatch order is generated and assigned to the operation and maintenance team closest to the faulty equipment, requiring them to arrive on-site within 1 hour; for Level II critical faults (such as transformer polarity errors or abnormal data from measurement and control devices), a high-priority maintenance dispatch order is generated and assigned to the corresponding professional operation and maintenance team, requiring them to arrive on-site within 4 hours; for Level III general faults (such as relay action delay slightly exceeding the threshold or fault recorder storing alarms), a regular maintenance dispatch order is generated and scheduled for completion within 24 hours. The dispatch order information includes the name of the faulty equipment, its location, fault type, severity level, handling suggestions, and a list of required tools and spare parts.

[0041] The advantages of this invention lie in its ability to dynamically adapt the big data model to complex scenarios such as secondary circuit signal fluctuations, message interference, and operating condition changes during equipment operation by collecting multi-dimensional operational data such as secondary circuit voltage, current, and message signals, combining factor analysis to extract core features, and then using an incremental model update mechanism. Compared to traditional fixed threshold detection methods, this method can adjust model parameters according to the real-time operating status of the equipment, improving adaptability to dynamic operating conditions by more than 60%. It effectively solves the problem of significant decrease in detection accuracy under complex operating conditions in traditional methods, exhibits outstanding anti-signal interference capabilities, high detection accuracy, reliable data, and accurate capture of early fault signals, achieving full-cycle detection, high automation, and improved operation and maintenance efficiency.

[0042] like Figure 2As shown, based on the above method, the present invention provides an electrical secondary equipment fault detection device, comprising: a data acquisition unit 201, used to acquire operating status information of electrical secondary equipment in the power grid and obtain acquired data; a training unit 202, used to train a big data model using the acquired data and obtain a trained model; a detection unit 203, used to perform fault detection on electrical secondary equipment in the current power grid according to the trained model and obtain detection results; and a processing unit 204, used to determine a fault handling plan and dispatch maintenance orders based on the detection results.

[0043] It should be understood that all relevant content of each step involved in the above method embodiments can be referenced to the functional description of the corresponding functional module, and will not be repeated here. Furthermore, the use of suffixes such as "module," "component," or "unit" to represent elements is merely for the purpose of illustrative purposes and has no specific meaning in itself. Therefore, "module," "component," or "unit" can be used interchangeably. Terminals can be implemented in various forms. For example, the terminals described in this invention may include mobile terminals such as mobile phones, tablets, laptops, handheld computers, personal digital assistants (PDAs), portable media players (PMPs), navigation devices, wearable devices, smart bracelets, pedometers, etc., as well as fixed terminals such as digital TVs and desktop computers. The following description will use mobile terminals as examples; those skilled in the art will understand that, in addition to elements specifically designed for mobile purposes, the construction according to embodiments of the present invention can also be applied to fixed-type terminals.

[0044] In other embodiments of the present invention, an electronic device 300 is disclosed, such as... Figure 3 As shown, the device may include: one or more processors 301; memory 302; display 303; one or more application programs (not shown); and one or more computer programs 304. These devices can be connected via one or more communication buses 305. The one or more computer programs 304 are stored in the memory 302 and configured to be executed by the one or more processors 301. The one or more computer programs 304 include instructions that can be used to perform actions such as... Figure 1 Each step in the corresponding embodiment.

[0045] Processor 301 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. A general-purpose processor can be a microprocessor or any conventional processor.

[0046] The memory 302 can be an internal storage unit of the electronic device 300, such as a hard disk or RAM of the electronic device 300. The memory 302 can also be an external storage device of the electronic device 300, such as a plug-in hard disk, Smart Media Card (SMC), Secure Digital (SD) card, or FlashCard equipped on the electronic device 300. Furthermore, the memory 302 can include both internal and external storage units of the electronic device 300. The memory 302 is used to store computer programs and other programs and data required by the electronic device. The memory 302 can also be used to temporarily store data that has been output or will be output.

[0047] The computer program 304 can be divided into one or more modules / units. The one or more modules / units can be a series of computer program instruction segments that can perform a specific function. The instruction segments are used to describe the execution process of the computer program 304 in the electronic device 300.

[0048] In addition to the above-described structure, those skilled in the art will understand that Figure 3 This is merely an example of electronic device 300 and does not constitute a limitation on electronic device 300. Electronic device 300 may include more or fewer components than shown, or combine certain components, or different components. For example, electronic device may also include input / output devices, network access devices, buses, etc.

[0049] 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 invention. 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.

[0050] Based on the above embodiments, the present invention also discloses a computer-readable storage medium having at least one computer program stored thereon, wherein the computer program, when executed by a processor, implements the methods described in the foregoing embodiments.

[0051] Those skilled in the art will understand that all or part of the steps in the methods of the above embodiments can be implemented by a program instructing a processor, and the program can be stored in a computer-readable storage medium. The storage medium is a non-transitory medium, such as random access memory, read-only memory, flash memory, hard disk, solid-state drive, magnetic tape, floppy disk, optical disk, and any combination thereof. The aforementioned storage medium can be any available medium accessible to a computer or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., digital video disc (DVD)), or a semiconductor medium (e.g., solid-state disk (SSD)).

[0052] The descriptions of the processes or structures corresponding to the above figures each have their own emphasis. For parts of a process or structure that are not described in detail, please refer to the relevant descriptions of other processes or structures.

[0053] In summary, the present invention discloses a method, apparatus, device, medium, and program for fault detection of electrical secondary equipment. By combining factor analysis with big data technology, it constructs a dynamically adaptable, highly anti-interference, accurate, and efficient fault detection model, enabling real-time fault detection of secondary equipment throughout its entire lifecycle. This effectively overcomes the limitations of traditional detection methods, mines the potential correlations between equipment operating parameters through a data-driven approach, accurately captures early fault signals, and adapts to parameter drift after long-term equipment operation, thus providing a reliable guarantee for the safe and stable operation of power grid secondary equipment.

[0054] Although the embodiments of the present invention have been described in detail above, it will be apparent to those skilled in the art that various modifications and variations can be made to these embodiments. The above descriptions are merely embodiments of the present invention and do not limit the patent scope of the present invention. However, it should be understood that such modifications and variations fall within the scope and spirit of the present invention. Moreover, the present invention described herein may have other embodiments and can be implemented or realized in various ways. All equivalent transformations made based on the description and drawings of the present invention, or direct or indirect applications in related technical fields, are similarly included within the patent protection scope of the present invention.

Claims

1. A method for fault detection in secondary electrical equipment, characterized in that, include: Collect operating status information of electrical secondary equipment in the power grid to obtain the collected data; The collected data is used to train the big data model to obtain a trained model; Based on the trained model, fault detection is performed on the electrical secondary equipment in the current power grid to obtain the detection results; Based on the test results, a fault handling plan is determined and a repair order is dispatched.

2. The method according to claim 1, characterized in that, Also includes: The collected data is preprocessed and dimensionality reduced. The preprocessing includes outlier handling, missing value handling, and standardization. Outlier handling uses the 3σ criterion to remove extreme biased data, and missing value handling uses interpolation to complete the data. The dimensionality reduction process includes calculating the correlation coefficient matrix of the standardized data to determine data suitability; performing eigenvalue decomposition on the correlation coefficient matrix to obtain the eigenvalues ​​of common factors; and selecting effective common factors based on the principle that the eigenvalues ​​are greater than 1. Orthogonally rotate the common factors to make their meaning clearer; construct factor vectors based on the selected common factors; and project the dataset onto the common factors to reduce dimensionality.

3. The method according to claim 1, characterized in that, The big data model is a big data-driven gradient boosting tree model. This model is built on the XGBoost framework and includes a feature selection layer, a gradient boosting layer, and an output layer. The feature selection layer uses a variance thresholding method to filter invalid features. The gradient boosting layer consists of three boosting units. The first boosting unit contains 64 decision tree nodes and uses L1 regularization to suppress overfitting. The second boosting unit contains 32 decision tree nodes and uses L2 regularization to optimize weights. The third boosting unit contains 16 decision tree nodes and uses cross-validation to optimize parameters. The outputs of each boosting unit are summed using a weighted fusion method.

4. The method according to claim 3, characterized in that, The output layer of the big data model is a multi-class output unit. It uses the cross-entropy loss function to quantify the difference between the predicted and actual values, and uses a parallel gradient descent optimizer to adjust the model parameters to meet the needs of efficient training with large amounts of data. During training, an early stopping strategy is used to prevent overfitting.

5. The method according to claim 1, characterized in that, The big data model is trained using the collected data to obtain a trained model by: dividing the collected data into a sample set and a test set, and training the big data model to obtain an intermediate prediction model for each type of electrical secondary equipment; based on the intermediate prediction model for each target equipment, incremental updates and parameter optimizations are performed in combination with real-time collected equipment operation data to obtain a prediction model adapted to different operating conditions of secondary equipment.

6. The method according to claim 1, characterized in that, The electrical secondary equipment includes relays, transformers, measurement and control devices, protection devices, and fault recorders; and / or the acquired data includes secondary circuit voltage, secondary circuit current, switch status, message signals, action sequence, insulation resistance, and signal amplitude.

7. An electrical secondary equipment fault detection device, used in the method according to any one of claims 1-6, characterized in that, include: The data acquisition unit is used to collect operating status information of electrical secondary equipment in the power grid and obtain the collected data. The training unit is used to train the big data model using the collected data to obtain a trained model. The detection unit is used to perform fault detection on the electrical secondary equipment in the current power grid according to the trained model, and obtain the detection results. The processing unit is used to determine a fault handling plan and dispatch maintenance orders based on the detection results.

8. An electronic device, characterized in that, The device includes a memory and a processor, wherein the memory stores a program that can run on the processor, and when the program is executed by the processor, causes the electronic device to perform the method of any one of claims 1-6.

9. A readable storage medium storing a program, characterized in that, When the program is executed, it implements the method of any one of claims 1-6.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the method of any one of claims 1-6.

Images