Electrical system relay protection fault diagnosis method and device and storage medium

Abstract

The invention discloses an electrical system relay protection fault diagnosis method and device and a storage medium, relates to the technical field of relay protection, and adopts the scheme that relay protection data of a power system are efficiently processed and analyzed by combining two technologies of U-MAP and kernel ridge regression. The method comprises the following steps: firstly, performing feature selection and data preprocessing by using U-MAP, and capturing a nonlinear structure and a complex mode in data; and then, classification prediction is carried out by using kernel ridge regression, so that the accuracy and response speed of relay protection are improved. The multi-model fusion method is greatly advanced in accuracy, recall rate and F1 value. A traditional method can judge based on a single electrical quantity threshold value, is easy to interfere and is low in performance. Therefore, various indexes of the multi-model fusion method are remarkably superior to those of a traditional method.

Description

Technical Field

[0001] The present application relates to the technical field of relay protection, and particularly relates to a method, device and storage medium for fault diagnosis of relay protection in an electrical system. Background Art

[0002] With the increasing complexity of the power grid structure, the requirements for the reliability of power grid operation are also continuously improving. Traditional relay protection methods mainly rely on fixed threshold settings and pre-set protection strategies. The defects of traditional relay protection methods are mainly reflected in poor flexibility, insufficient adaptability, low intelligent level, weak anti-interference ability, etc. With the development of the power system towards intelligence and digitization, traditional relay protection methods have gradually been unable to meet the needs of modern power systems.

[0003] Therefore, there is an urgent need to explore a more intelligent and efficient new relay protection method to meet the actual needs of modern power systems. Summary of the Invention

[0004] The present application provides a method, device and storage medium for fault diagnosis of relay protection in an electrical system to solve the above problems. The solution efficiently processes and analyzes the relay protection data of the power system by combining two technologies, U-MAP and kernel ridge regression. First, U-MAP is used for feature selection and data preprocessing to capture the non-linear structure and complex patterns in the data; then, kernel ridge regression is used for classification prediction to improve the accuracy and response speed of relay protection.

[0005] On the one hand, the present application provides a method for fault diagnosis of relay protection in an electrical system, and the method includes the following steps:

[0006] Step S1: Collect the fault signal data of the relay protection equipment in the power system, including: the time-domain characteristics of current, voltage, and power;

[0007] Step S2: Clean the fault signal data, identify and process the noise data and missing values;

[0008] Step S3: Use the U-MAP algorithm to perform dimensionality reduction processing on the cleaned data, map the high-dimensional data to a low-dimensional space, and extract non-linear topological features;

[0009] Step S4: Construct a kernel ridge regression classification model, input the dimensionality-reduced feature data, and perform fault type classification prediction;

[0010] Step S5: Trigger corresponding relay protection actions according to the classification results.

[0011] In one implementation of the present application, in step S3, the dimensionality reduction process of the U-MAP algorithm includes: retaining the local topological structure and complex patterns of high-dimensional data through manifold learning, mapping it to a low-dimensional space, and reducing data redundancy and computational complexity.

[0012] In one implementation of the present application, in step S4, the construction of the kernel ridge regression classification model includes: using a Gaussian radial basis kernel function to map low-dimensional non-linear features to a high-dimensional linearly separable space, and combining the regularization parameter λ to balance the model fitting and generalization capabilities.

[0013] In one implementation of the present application, in step S2, the method for identifying noise data is: according to the threshold range of electrical quantities during the normal operation of the power system, marking and correcting abnormal data that exceeds the threshold, specifically including current peak value, voltage instantaneous value, and harmonic content.

[0014] In one implementation of the present application, in step S4, the training of the kernel ridge regression model includes: using the historical fault data set, inputting label data including short-circuit faults, open-circuit faults, and equipment overheating faults, and learning the mapping relationship between input features and fault types.

[0015] In one implementation of the present application, the low-dimensional features after dimensionality reduction in step S3 include: current peak value, voltage instantaneous value, active power, power factor, harmonic content, voltage unbalance degree, and frequency deviation.

[0016] In one implementation of the present application, the trigger logic for the relay protection action in step S5 is: according to the fault type probability output by the kernel ridge regression model, matching the preset protection strategy library, and preferentially triggering the protection mechanism corresponding to the fault type with the highest probability.

[0017] In one implementation of the present application, the method further includes: updating the kernel ridge regression model through real-time collected electrical quantity data, and optimizing the model parameters using the incremental learning algorithm to adapt to the dynamic changes of the power system.

[0018] The present application also provides an electrical system relay protection fault diagnosis device, the device includes: at least one processor; and, a memory communicatively connected to the at least one processor; wherein, the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can complete the foregoing electrical system relay protection fault diagnosis method.

[0019] The present application also provides a non-volatile computer storage medium for fault diagnosis of electrical system relay protection, storing computer-executable instructions, and the computer-executable instructions are executed by a processor to implement the foregoing method for fault diagnosis of electrical system relay protection.

[0020] The method, device and storage medium for fault diagnosis of electrical system relay protection provided by the present application have the following beneficial effects:

[0021] (1) Deeply integrate the two models of non-linear dimensionality reduction and kernel ridge regression, and give full play to their respective advantages: retain the local topological structure of high-dimensional data, and avoid the problem of information loss of traditional dimensionality reduction methods (such as PCA). And complex non-linear features can be mapped through kernel functions to achieve accurate classification.

[0022] (2) Have efficient data preprocessing capabilities. U-MAP maps high-dimensional electrical quantities (current, voltage, power, etc.) to a low-dimensional space, reducing the computational complexity while retaining key features.

[0023] (3) Have non-linear classification performance: Kernel ridge regression combines with Gaussian radial basis kernel function (RBF) to transform low-dimensional non-linear problems into high-dimensional linearly separable problems, and accurately identify complex fault patterns; the regularization parameter (λ) balances the model fitting and generalization capabilities, avoids overfitting, and improves the adaptability to unknown faults. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] The drawings described herein are used to provide a further understanding of the present application, and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation to the present application. In the drawings:

[0025] Figure 1 is a flowchart of a method for fault diagnosis of electrical system relay protection provided by an embodiment of the present application;

[0026] Figure 2 is a schematic diagram of a device for fault diagnosis of electrical system relay protection provided by an embodiment of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0027] To make the objectives, technical solutions and advantages of the present application clearer, the technical solutions of the present application will be clearly and completely described below in conjunction with specific embodiments of the present application and the corresponding drawings. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present application.

[0028] The embodiments of the present application provide a method, device, and storage medium for relay protection fault diagnosis of an electrical system. Specifically, in the first aspect, a method for relay protection fault signal feature selection and data preprocessing based on U-MAP is first designed. This method can effectively capture the non-linear structure and complex patterns in the data when processing non-linear and time-varying relay protection data. In the second aspect, a relay protection method based on kernel ridge regression is also designed. This method classifies by using kernel ridge regression technology through the analysis of the fault feature data of the power system, thereby improving the accuracy and response speed of relay protection. The technical solutions proposed in the embodiments of the present application are described in detail below with reference to the accompanying drawings.

[0029] Figure 1 It is a flowchart of a method for relay protection fault diagnosis of an electrical system provided by the embodiments of the present application. As Figure 1 shown, the method mainly includes the following steps:

[0030] Step S1: Collect the fault signal data of the relay protection equipment of the power system, including: the time-domain characteristics of current, voltage, and power;

[0031] Step S2: Clean the fault signal data, identify and process the noise data and missing values;

[0032] Step S3: Use the U-MAP algorithm to perform dimensionality reduction processing on the cleaned data, map the high-dimensional data to a low-dimensional space, and extract non-linear topological features;

[0033] Step S4: Construct a kernel ridge regression classification model, input the dimensionality-reduced feature data, and perform fault type classification prediction;

[0034] Step S5: Trigger corresponding relay protection actions according to the classification results.

[0035] In the embodiments of this application, as an advanced manifold learning algorithm, U-MAP can more effectively preserve the topological structure and local features of data compared with traditional dimensionality reduction methods such as principal component analysis (PCA). Through the dimensionality reduction mapping of high-dimensional fault signal data (such as multi-dimensional electrical quantity information including current, voltage, power, etc.), it is transformed into a low-dimensional space, which not only reduces the data storage and calculation costs, but also improves the effectiveness and redundancy of features, enabling subsequent model training to focus on key feature information and laying a solid foundation for accurate fault diagnosis. It is mainly divided into the following steps: a. Collect original data and extract effective features. These features include the instantaneous values and time-domain features of electrical quantities such as current, voltage, and power. b. Identify and process noise data. The method of identifying noise data usually adopts a threshold-based approach. A reasonable threshold is set according to the value range of electrical quantities during the normal operation of the power system. When the collected data exceeds this threshold range, it is initially determined as noise data. c. Based on the research data of the previous two steps, for missing and abnormal data, the linear interpolation method is used. According to the known data points before and after the missing value, the missing value is estimated according to the linear relationship.

[0036] In the embodiments of this application, the kernel ridge regression model accurately judges the fault type based on its powerful non-linear classification and prediction ability. During the training stage of the kernel ridge regression model, through learning a large number of sample data with real fault labels, a mapping relationship from input features to fault types is constructed. In practical applications, in the face of new input feature data, the model quickly calculates and outputs the prediction probabilities of each fault type, and determines the most likely fault type according to the probability size. For example, in a training set containing samples of various short-circuit faults, open-circuit faults, and equipment overheating faults, etc., the kernel ridge regression model learns the subtle feature differences in the changes of electrical quantities such as current, voltage, and power under different fault types. When the measured electrical quantity feature data at a certain moment is input, the model can accurately judge whether the fault belongs to three-phase short circuit, two-phase short circuit, or other specific types, providing a key basis for the accurate triggering of subsequent relay protection actions.

[0037] The working process of the relay protection method based on kernel ridge regression is as follows: a. Input the feature data after U-MAP dimensionality reduction and correction to ensure that the data dimension is appropriate and outliers have been corrected. b. Perform data standardization processing to eliminate the influence of different feature dimensions. For example, normalize electrical quantities such as current and voltage. c. Use historical data to train the kernel ridge regression model. Generate a regression model. d. Input the preprocessed new sample features into the model, calculate the kernel function values with the training data, and generate a prediction output. e. According to the prediction output and the fault classification result, match the preset protection strategy. The comparison of the accuracy rate and recall rate between the traditional method and the method of this application embodiment is shown in Table 1 below.

[0038] Table 1 Comparison table of accuracy rate and recall rate between traditional method and this method

[0039]

[0040] The multi - model fusion method significantly leads in terms of accuracy, recall rate, and F1 - value. From the table, traditional methods can make judgments based on a single electrical quantity threshold, are vulnerable to interference, and have low performance. Therefore, all indicators of the multi - model fusion method are significantly better than those of traditional methods.

[0041] The above is a relay protection fault diagnosis method for an electrical system provided by an embodiment of the present application. Based on the same inventive concept, an embodiment of the present application also provides a relay protection fault diagnosis device for an electrical system. Figure 2 Schematic diagram of a relay protection fault diagnosis device for an electrical system provided by an embodiment of the present application, as Figure 2 shown, the device mainly includes: at least one processor 201; and a memory 202 communicatively connected to the at least one processor; wherein, the memory 202 stores instructions executable by the at least one processor 201, and the instructions are executed by the at least one processor 201 to enable the at least one processor 201 to complete the aforementioned relay protection fault diagnosis method for an electrical system.

[0042] In addition, an embodiment of the present application also provides a non - volatile computer storage medium for relay protection fault diagnosis of an electrical system, storing computer - executable instructions, and the computer - executable instructions are executed by a processor to implement the aforementioned relay protection fault diagnosis method for an electrical system.

[0043] The present invention is described with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, and 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, such that the instructions executed by the processor of the computer or other programmable data - processing devices generate means for implementing the functions specified in Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0044] These computer program instructions can also be stored in a computer - readable memory capable of guiding a computer or other programmable data - processing devices to work in a specific manner, such that the instructions stored in the computer - readable memory generate a manufactured product including instruction means, and the instruction means implements the functions specified in Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0045] These computer program instructions can also be loaded onto a computer or other programmable data processing apparatus, so that a series of operation steps are executed on the computer or other programmable apparatus to produce a computer-implemented process, and thus the instructions executed on the computer or other programmable apparatus provide steps for realizing the functions specified in one process Figure 1 one process or multiple processes and / or blocks Figure 1 or steps for realizing the functions specified in multiple blocks.

[0046] In a typical configuration, a computing device includes one or more processors (CPUs), an input / output interface, a network interface, and a memory.

[0047] Each embodiment in this application is described in a progressive manner. For the same or similar parts among the embodiments, reference can be made to each other. Each embodiment focuses on the differences from other embodiments. In particular, for the apparatus embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can refer to the description of the method embodiment.

[0048] It should also be noted that the term "comprising", "including" or any other variant thereof is intended to cover a non-exclusive inclusion, so that a process, method, commodity or device including a series of elements not only includes those elements but also includes other elements not explicitly listed, or further includes elements inherent to such process, method, commodity or device. Without more limitations, the element defined by the statement "including one..." does not exclude the existence of additional identical elements in the process, method, commodity or device including the said element.

[0049] The above are only the embodiments of the present application and are not intended to limit the present application. For those skilled in the art, various changes and modifications 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.

Claims

1. A method for diagnosing relay protection faults in an electrical system, characterized in that: The method comprises the following steps: Step S1: Collecting fault signal data of the power system relay protection equipment, including: time domain characteristics of current, voltage, and power; Step S2: performing data cleaning on the fault signal data, identifying and processing noise data and missing values; Step S3: Use the U-MAP algorithm to perform dimensionality reduction processing on the cleaned data, map the high-dimensional data to a low-dimensional space, and extract nonlinear topological features; Step S4: construct a kernel ridge regression classification model, input the feature data after dimension reduction, and perform fault type classification prediction; Step S5: triggering corresponding relay protection actions according to the classification results.

2. The electrical system relay protection fault diagnosis method according to claim 1, characterized in that: In step S3, the dimension reduction processing of the U-MAP algorithm includes: retaining the local topological structure and complex patterns of high-dimensional data through manifold learning, mapping to low-dimensional space, and reducing data redundancy and computational complexity.

3. The method for diagnosing relay protection faults in an electrical system according to claim 1, characterized in that: The construction of the kernel ridge regression classification model in step S4 includes: using a Gaussian radial basis kernel function to map low-dimensional nonlinear features to a high-dimensional linearly separable space, and combining a regularization parameter λ to balance model fitting and generalization capabilities.

4. The method for diagnosing relay protection faults in an electrical system according to claim 1, characterized in that: In step S2, the noise data identification method is: according to the threshold range of electrical quantities during normal operation of the power system, abnormal data exceeding the threshold is marked and corrected, specifically including current peak value, voltage instantaneous value and harmonic content.

5. The electrical system relay protection fault diagnosis method according to claim 1, characterized in that: In step S4, the training of the kernel ridge regression model includes: using the historical fault data set, inputting label data including short circuit faults, open circuit faults and equipment overheating faults, and learning the mapping relationship between input features and fault types.

6. The method for diagnosing relay protection faults in an electrical system according to claim 1, characterized in that: The low-dimensional features after dimension reduction in step S3 include: current peak value, voltage instantaneous value, active power, power factor, harmonic content, voltage imbalance and frequency deviation.

7. The electrical system relay protection fault diagnosis method according to claim 1, characterized in that: The triggering logic of the relay protection action in step S5 is: according to the fault type probability output by the kernel ridge regression model, matching the preset protection strategy library, and preferentially triggering the protection mechanism corresponding to the fault type with the highest probability.

8. The method for diagnosing relay protection faults in an electrical system according to claim 1, characterized in that: The method also includes: updating the kernel ridge regression model through the real-time collected electrical quantity data, optimizing the model parameters by using the incremental learning algorithm, and adapting to the dynamic changes of the power system.

9. An electrical system relay protection fault diagnosis device, characterized in that: The device comprises: at least one processor; and, a memory communicatively connected to the at least one processor; wherein, The memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can complete the electrical system relay protection fault diagnosis method described in any one of claims 1-8.

10. A non-volatile computer storage medium for diagnosing relay protection faults in an electrical system, storing computer executable instructions, characterized in that: The computer executable instructions are executed by a processor to implement an electrical system relay protection fault diagnosis method as described in any one of claims 1-8.

Images