Fault Analysis Method for Power Transmission and Transformation Control System

Abstract

The present invention discloses a fault analysis method for a power transmission and transformation control system, which relates to the technical field of power system fault diagnosis. The method comprises collecting operating data from the power transmission and transformation control system, grouping and labeling the data, and forming a time series data matrix; using a variational mode decomposition method to adaptively extract components from the time series data matrix, obtain intrinsic mode components, construct a feature space containing multi-dimensional features, and extract fault feature points; calculating the similarity between the fault feature points and historical fault features in a fault sample library to generate a fault correlation topology graph G, wherein the node correlation strength in the fault correlation topology graph G is used to classify the fault severity; and retrieving the corresponding treatment plan from a preset fault handling strategy library based on the fault severity, and outputting the corresponding fault cause analysis results. The present invention achieves significant improvements in fault feature extraction accuracy, fault location accuracy, and treatment plan matching efficiency, and has important engineering application value.

Description

Technical Field

[0001] The present invention relates to the technical field of power system fault diagnosis, and in particular to a fault analysis method for a power transmission and transformation control system. Background Art

[0002] As the nerve center of the power system, the safe and stable operation of the power transmission and transformation control system is crucial for power production and grid dispatch. Traditional fault analysis methods rely primarily on empirical judgment and manual inspections, which significantly limits their accuracy and real-time performance. With the rapid development of technologies such as big data and artificial intelligence, fault diagnosis is gradually evolving towards intelligent and precise methods, but existing technologies still face numerous bottlenecks.

[0003] Existing fault analysis technologies for power transmission and transformation control systems commonly suffer from inaccurate feature extraction, limited correlation analysis, and significant fault prediction lag. These technologies often struggle to effectively process complex, multi-dimensional, and nonlinear operating data and lack the ability to dynamically track fault evolution. Traditional methods are particularly inadequate in identification accuracy and response speed when faced with sudden and hidden faults. Traditional methods typically rely on a single or limited feature dimension and fail to fully capture the inherent laws of system operation, resulting in significant blind spots and lags in fault diagnosis.

[0004] In view of the limitations of existing power transmission and transformation control system fault analysis technology, the present invention proposes a fault analysis method for power transmission and transformation control system, aiming to solve the problems of inaccurate feature extraction and single correlation analysis in traditional fault diagnosis technology. Summary of the Invention

[0005] In view of the problems of inaccurate feature extraction, single correlation analysis and delayed fault prediction in the existing power transmission and transformation control system fault analysis technology, the present invention is proposed.

[0006] Therefore, the problem to be solved by the present invention is how to accurately extract multi-dimensional fault features through adaptive modal decomposition technology to achieve accurate fault identification, correlation analysis and rapid response.

[0007] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0008] In a first aspect, an embodiment of the present invention provides a fault analysis method for a power transmission and transformation control system, comprising: collecting operating data in the power transmission and transformation control system, and grouping and labeling the operating data to form a time series data matrix; using a variational mode decomposition method to adaptively extract components of the time series data matrix to obtain inherent mode components; based on the inherent mode components, constructing a feature space containing multi-dimensional features to extract fault feature points; calculating the similarity between the fault feature points and historical fault features in a fault sample library to generate a fault association topology graph G, wherein the node association strength in the fault association topology graph G grades the fault degree; retrieving a corresponding processing solution from a preset fault processing strategy library according to the fault level, and outputting a corresponding fault cause analysis result.

[0009] As a preferred solution of the fault analysis method for power transmission and transformation control system of the present invention, the method for obtaining the fault cause analysis result is:

[0010] Establish a mapping relationship table M between fault levels and processing solutions in a preset fault handling strategy library, wherein the mapping relationship table M includes the fault device type, fault location, fault level, processing priority P, and response time window T;

[0011] According to the node importance index Wi in the fault association topology graph G, a set of processing solutions S corresponding to the fault level in the mapping relationship table M is retrieved, wherein each solution in the set of processing solutions S includes a description of the fault cause, a sequence of treatment measures, and maintenance resource requirements;

[0012] Calculate the matching score D for each solution in the solution set S i , the specific formula is as follows:

[0013]

[0014] Among them, D i is the matching score of fault handling solution i, λ1 is the device matching weight factor, μ e is the device matching degree, λ2 is the position matching weight factor, μ p is the position matching degree, λ3 is the historical success rate weight factor, η is the historical processing success rate, Imp is the impact factor, ω u is the weight coefficient of node u, Cor() is the correlation function, S j is the success probability of the jth processing sub-strategy, V is the total number of network nodes, and r is the total number of processing sub-strategies;

[0015] The processing solution with the highest matching score is selected as the optimal solution, and an analysis report R is generated. The solutions are sorted according to the processing priority P, and the response time window T of each treatment measure is marked. The analysis report R includes the fault cause chain, key influencing factors and treatment suggestions.

[0016] As a preferred solution of the fault analysis method for power transmission and transformation control system of the present invention, the method for generating the fault correlation topology graph G is to calculate the similarity value between the feature vectors of the fault feature points and the historical fault features in the fault sample library using the cosine similarity function. The specific formula is as follows:

[0017]

[0018] Among them, Q ab is the similarity value between fault feature point a and historical fault feature point b, x ak is the kth component of the ath eigenvector, y bk is the kth component of the bth eigenvector, m is the feature dimension, and f() is the distance decay function;

[0019] According to the similarity value Q ab Construct the adjacency matrix A, when the similarity value Q ab When the similarity value is greater than the similarity threshold θ, the corresponding position in the adjacency matrix A is marked as 1; when the similarity value Q ab When the similarity threshold θ is less than or equal to the value of the similarity, the corresponding position in the adjacency matrix A is marked as 0; the adjacency matrix A is used to generate a fault correlation topology graph G, where the nodes of the fault correlation topology graph G represent the fault feature points, the edges represent the association relationship between the nodes, and the weight of the edge is set to the corresponding similarity value Q ab ;

[0020] Calculate the degree centrality C1 and betweenness centrality C2 of the nodes in the fault association topology graph G, and combine the similarity value Q ab Calculate the node importance index W i , the specific formula is as follows:

[0021]

[0022] Among them, W i is the importance index of node i, ρ is the degree centrality weight coefficient, C1(v i ) is the degree centrality of node i, C2(v i ) is the betweenness centrality of node i, τ is the betweenness centrality weight coefficient, φ is the similarity weight coefficient, and V is the total number of network nodes;

[0023] According to the node importance index W iThe fault severity is divided into level one fault, level two fault, level three fault and level four fault.

[0024] As a preferred solution of the fault analysis method for power transmission and transformation control system described in the present invention, wherein: the method for extracting the fault feature point is to calculate the time domain eigenvalue and frequency domain eigenvalue of each inherent modal component {IMF1, IMF2, ..., IMF3} respectively; the time domain eigenvalue and the frequency domain eigenvalue are combined into a feature vector F, and normalized; a Y-dimensional feature space is constructed according to the processed feature vector F, wherein each coordinate axis in the Y-dimensional feature space corresponds to a feature component in the feature vector F; based on the Y-dimensional feature space, data points with similar eigenvalues ​​are clustered to form feature clusters, and the local density and relative distance of each feature cluster are calculated using a density peak clustering algorithm; based on the product of the local density and the relative distance, the feature cluster center with the maximum product value is selected as the fault feature point.

[0025] As a preferred solution of the fault analysis method for power transmission and transformation control system described in the present invention, the method for obtaining the inherent modal components is to perform a variational modal decomposition operation on each type of operating data in the time series data matrix, decompose each time series signal into K inherent modal components, and calculate the center frequency and bandwidth constraints; through secondary optimization iterative calculation, the bandwidth constraints of adjacent inherent modal components are minimized. When the number of iterations reaches a preset value R or the bandwidth change is less than a threshold, the iteration is stopped, and the inherent modal components are sorted from high to low according to the center frequency to form a component sequence {IMF1, IMF2, ..., IMF3}.

[0026] As a preferred solution of the fault analysis method for a power transmission and transformation control system described in the present invention, the method for obtaining the time series data matrix is ​​as follows: the operating data in the power transmission and transformation control system is collected, wherein the operating data includes primary equipment data, secondary equipment data and environmental monitoring data; the operating data is timestamped according to the sampling time points, and the operating data with the same timestamp are combined along the time axis with the sampling period T as the unit; an M×N time series data matrix is ​​established for the annotated operating data, wherein the time series data matrix includes normal operating data and fault operating data, M represents the number of sampling time points, and N represents the number of types of operating data.

[0027] As a preferred solution of the fault analysis method for power transmission and transformation control system described in the present invention, the primary equipment data includes voltage value, current value, active power value and reactive power value; the secondary equipment data includes protection action signal, alarm signal and switch status signal; the environmental monitoring data includes temperature value, humidity value and vibration value.

[0028] In a second aspect, an embodiment of the present invention provides a fault analysis system for a power transmission and transformation control system, comprising: an acquisition and annotation module for acquiring operating data in the power transmission and transformation control system, and grouping and annotating the operating data to form a time series data matrix; an extraction module for adaptively extracting components from the time series data matrix using a variational mode decomposition method to obtain inherent mode components; a construction module for constructing a feature space containing multi-dimensional features based on the inherent mode components to extract fault feature points; a generation module for calculating the similarity between the fault feature points and historical fault features in a fault sample library to generate a fault association topology graph G, wherein the node association strength in the fault association topology graph G grades the fault degree; an analysis module for retrieving a corresponding processing solution from a preset fault processing strategy library according to the fault level, and outputting a corresponding fault cause analysis result.

[0029] In a third aspect, an embodiment of the present invention provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: when the computer program instructions are executed by the processor, the steps of the fault analysis method for a power transmission and transformation control system as described in the first aspect of the present invention are implemented.

[0030] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program instructions are executed by a processor, the steps of the fault analysis method for a power transmission and transformation control system as described in the first aspect of the present invention are implemented.

[0031] The beneficial effects of the present invention are as follows: the present invention realizes the unified management of multi-source heterogeneous data through the construction of a time series data matrix, solving the problem of chaotic data structure in traditional methods; adopts an improved variational mode decomposition method for adaptive component extraction, effectively overcoming the technical bottleneck of modal aliasing in traditional signal processing, and improving the accuracy of feature extraction; constructs a multidimensional feature space based on inherent modal components, breaking through the limitations of single feature expression; realizes the visualization expression of fault propagation paths through the construction of a fault correlation topology map, and improves the accuracy of fault location; adopts an intelligent matching mechanism to realize the automatic selection of processing solutions, and shortens the solution matching time. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] To more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. Those skilled in the art can also derive other drawings based on these drawings without inventive effort. Among them:

[0033] Figure 1 This is a flow chart of the fault analysis method for a power transmission and transformation control system according to Example 1. DETAILED DESCRIPTION

[0034] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0035] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0036] Secondly, the term "one embodiment" or "embodiment" herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in various places throughout this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive of other embodiments.

[0037] Example 1

[0038] Reference Figure 1 , which is the first embodiment of the present invention, provides a fault analysis method for a power transmission and transformation control system, comprising:

[0039] S1: Collecting operating data from the power transmission and transformation control system, and grouping and labeling the operating data to form a time series data matrix.

[0040] Specifically, the method for acquiring the time series data matrix is ​​to collect operating data in the power transmission and transformation control system, where the operating data includes primary equipment data, secondary equipment data and environmental monitoring data.

[0041] It should be noted that primary equipment data includes voltage values, current values, active power values ​​and reactive power values; secondary equipment data includes protection action signals, alarm signals and switch status signals, and environmental monitoring data includes temperature values, humidity values ​​and vibration values.

[0042] Furthermore, the operation data is timestamped according to the sampling time point, and the operation data with the same timestamp are combined along the time axis with the sampling period T as the unit, where the value range of T is 1ms to 100ms.

[0043] Furthermore, an M×N time series data matrix is ​​established for the labeled operating data, where M represents the number of sampling time points and N represents the number of types of operating data. The time series data matrix includes normal operating data and fault operating data, and each element in the matrix corresponds to a data sampling value.

[0044] S2: Adopting the variational mode decomposition method to perform adaptive component extraction on the time series data matrix to obtain the intrinsic mode components.

[0045] Specifically, the method for obtaining the intrinsic modal components is to perform a variational modal decomposition operation on each type of operating data in the time series data matrix, decompose each time series signal into K intrinsic modal components, and calculate the center frequency and bandwidth constraints, where the K value is dynamically adjusted by the penalty factor α, and the value range of α is 100 to 2000.

[0046] Furthermore, the specific formulas for center frequency and bandwidth constraints are as follows:

[0047]

[0048] Among them, τ k is the bandwidth constraint of the kth component, δ k is the center frequency of the kth component, is the p-th data point of the k-th component, P is the total number of data points, and k is the number of decomposed modal components.

[0049] Furthermore, the bandwidth constraints of adjacent natural mode components are minimized through secondary optimization iterative calculation. When the number of iterations reaches a preset value R or the bandwidth change is less than a threshold, the iteration is stopped and the natural mode components are sorted from high to low according to the center frequency to form a component sequence {IMF1, IMF2, ..., IMF3}.

[0050] It should be noted that the preset value R ranges from 500 to 1000, and the threshold value is 0.001; the intrinsic modal component reflects the signal characteristics of different frequency scales in the time series data matrix.

[0051] S3: Based on the inherent modal components, a feature space including multi-dimensional features is constructed to extract fault feature points.

[0052] Specifically, the method for extracting fault feature points is to calculate the time domain eigenvalue and the frequency domain eigenvalue of each intrinsic modal component {IMF1, IMF2, ..., IMF3} respectively.

[0053] It should be noted that the time domain eigenvalues ​​include mean, standard deviation, kurtosis and skewness; the frequency domain eigenvalues ​​include frequency center, frequency standard deviation and frequency band energy.

[0054] Furthermore, the time domain eigenvalues ​​and the frequency domain eigenvalues ​​are combined into a eigenvector F and normalized; a Y-dimensional feature space is constructed based on the processed eigenvector F, wherein each coordinate axis in the Y-dimensional feature space corresponds to a eigencomponent in the eigenvector F; based on the Y-dimensional feature space, data points with similar eigenvalues ​​are clustered to form feature clusters, and a density peak clustering algorithm is used to calculate the local density and relative distance of each feature cluster; based on the product of the local density and the relative distance, the center of the feature cluster with the maximum product value is selected as the fault feature point.

[0055] Furthermore, the specific formulas for local density and relative distance are as follows:

[0056]

[0057] Among them, Z i is the local density of the i-th data point, d ij is the Euclidean distance between data point i and data point j, d c is the cutoff distance, B is the total number of data points, and χ() is the cutoff function.

[0058] S4: Calculate similarity between the fault feature point and historical fault features in the fault sample library to generate a fault correlation topology graph G, wherein the node correlation strength in the fault correlation topology graph G is used to classify the fault degree.

[0059] Specifically, the method for generating the fault correlation topology graph G is to calculate the similarity value between the feature vectors of the fault feature points and the historical fault features in the fault sample library using the cosine similarity function. The specific formula is as follows:

[0060]

[0061] Among them, Q ab is the similarity value between fault feature point a and historical fault feature point b, x ak is the kth component of the ath eigenvector, y bk is the kth component of the bth eigenvector, m is the feature dimension, and f() is the distance decay function.

[0062] Furthermore, according to the similarity value Q ab Construct the adjacency matrix A, when the similarity value Q ab When the similarity value is greater than the similarity threshold θ, the corresponding position in the adjacency matrix A is marked as 1; when the similarity value Q ab When it is less than or equal to the similarity threshold θ, the corresponding position in the adjacency matrix A is marked as 0. The value range of the threshold θθ is 0.75 to 0.95.

[0063] Furthermore, the adjacency matrix A is used to generate a fault correlation topology graph G, wherein the nodes of the fault correlation topology graph G represent fault feature points, the edges represent the association relationship between nodes, and the weights of the edges are set to the corresponding similarity values ​​Q ab Calculate the degree centrality C1 and betweenness centrality C2 of the nodes in the fault association topology graph G, and combine the similarity value Q ab Calculate the node importance index W i , the specific formula is as follows:

[0064]

[0065] Among them, W i is the importance index of node i, ρ is the degree centrality weight coefficient, C1(v i ) is the degree centrality of node i, C2(v i ) is the betweenness centrality of node i, τ is the betweenness centrality weight coefficient, φ is the similarity weight coefficient, and V is the total number of network nodes;

[0066] Specifically, according to the node importance index W i The fault severity is divided into level one fault, level two fault, level three fault and level four fault.

[0067] It should be noted that when the node importance index W i When ≥0.85, it is determined to be a level 1 fault and needs to be shut down immediately. The system will issue a red warning signal, requiring the emergency response team to arrive at the scene within 15 minutes and activate the highest response level of the emergency plan. When 0.65≤node importance index W i When the value is less than 0.85, it is considered a secondary fault and needs to be shut down within 2 hours. The system will issue an orange warning signal and maintenance personnel should arrive at the site within 30 minutes and prepare backup equipment and emergency plans. When the value is less than 0.45, the node importance index W i When the node importance index W is less than 0.65, it is judged as a level 3 fault and can be handled before the end of the work shift. The system issues a yellow warning signal and maintenance personnel are required to conduct an inspection within 2 hours and closely monitor the equipment operation status. i When it is <0.45, it is determined to be a level 4 fault, which can be handled during routine maintenance. The system will issue a blue prompt signal, record the fault information and arrange a routine maintenance plan.

[0068] Furthermore, a dynamic adjustment mechanism for fault levels in special circumstances is set: when a device triggers three or more faults consecutively within 24 hours, the fault level is automatically increased by one, a frequent fault warning is recorded, and equipment reliability analysis is started; when multiple devices fail at the same time and the correlation degree exceeds 0.7, the highest level is taken as the system fault level, and the group fault analysis mode is started; when a fault occurs at a critical node in the system, the fault level is automatically increased by one level, and the backup system switching procedure is started; when the scope of the fault continues to expand, the fault level is re-evaluated every 10 minutes, the emergency response level is dynamically adjusted, and the fault handling resource allocation is expanded.

[0069] S5: Retrieve the corresponding processing solution from the preset fault processing strategy library according to the fault level, and output the corresponding fault cause analysis result.

[0070] Specifically, the method for obtaining the fault cause analysis results is to establish a mapping relationship table M between fault levels and processing solutions in a preset fault handling strategy library, wherein the mapping relationship table M includes the fault device type, fault location, fault level, processing priority P and response time window T.

[0071] Furthermore, according to the node importance index Wi in the fault association topology graph G, a set of processing solutions S corresponding to the fault level in the mapping relationship table M is retrieved, wherein each solution in the processing solution set S includes a description of the fault cause, a sequence of treatment measures, and maintenance resource requirements; a matching score D is calculated for each solution in the processing solution set S. i , the specific formula is as follows:

[0072]

[0073] Among them, D i is the matching score of fault handling solution i, λ1 is the device matching weight factor, μ e is the device matching degree, λ2 is the position matching weight factor, μ p is the position matching degree, λ3 is the historical success rate weight factor, η is the historical processing success rate, Imp is the impact factor, ω u is the weight coefficient of node u, Cor() is the correlation function, S j is the success probability of the jth processing sub-strategy, V is the total number of network nodes, and r is the total number of processing sub-strategies.

[0074] Furthermore, the processing solution with the highest matching score is selected as the optimal solution, and an analysis report R is generated. The solutions are sorted according to the processing priority P, and the response time window T of each treatment measure is marked. The analysis report R includes the fault cause chain, key influencing factors and treatment suggestions.

[0075] Furthermore, this embodiment also provides a fault analysis system for a power transmission and transformation control system, including: an acquisition and annotation module for acquiring operating data in the power transmission and transformation control system, and grouping and annotating the operating data to form a time series data matrix; an extraction module for adaptively extracting components from the time series data matrix using a variational mode decomposition method to obtain inherent mode components; a construction module for constructing a feature space containing multi-dimensional features based on the inherent mode components to extract fault feature points; a generation module for calculating the similarity between the fault feature points and historical fault features in a fault sample library to generate a fault association topology graph G, wherein the node association strength in the fault association topology graph G grades the fault degree; an analysis module for retrieving a corresponding processing solution from a preset fault processing strategy library according to the fault level, and outputting a corresponding fault cause analysis result.

[0076] This embodiment also provides a computer device suitable for use in a fault analysis method for a power transmission and transformation control system, comprising a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute computer-executable instructions to implement the fault analysis method for a power transmission and transformation control system proposed in the above embodiment.

[0077] The computer device may be a terminal, comprising a processor, a memory, a communication interface, a display screen and an input device connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device comprises a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner may be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. The display screen of the computer device may be a liquid crystal display or an electronic ink display screen, and the input device of the computer device may be a touch layer covering the display screen, or a button, trackball or touchpad provided on the housing of the computer device, or an external keyboard, touchpad or mouse.

[0078] This embodiment also provides a storage medium having a computer program stored thereon, which, when executed by a processor, implements the following steps: collecting operating data from a power transmission and transformation control system, and grouping and annotating the operating data to form a time series data matrix; using a variational mode decomposition method to adaptively extract components from the time series data matrix to obtain inherent mode components; based on the inherent mode components, constructing a feature space containing multi-dimensional features to extract fault feature points; calculating the similarity between the fault feature points and historical fault features in a fault sample library to generate a fault association topology graph G, wherein the node association strength in the fault association topology graph G classifies the degree of fault; retrieving a corresponding processing solution from a preset fault processing strategy library according to the fault level, and outputting a corresponding fault cause analysis result.

[0079] In summary, the present invention realizes the unified management of multi-source heterogeneous data by constructing a time series data matrix, solving the problem of chaotic data structure in traditional methods; adopts an improved variational mode decomposition method for adaptive component extraction, effectively overcoming the technical bottleneck of modal aliasing in traditional signal processing and improving the accuracy of feature extraction; constructs a multidimensional feature space based on the inherent modal components, breaking through the limitations of single feature expression; realizes the visualization expression of the fault propagation path through the construction of a fault correlation topology map, and improves the accuracy of fault location; adopts an intelligent matching mechanism to realize the automatic selection of processing solutions and shorten the solution matching time.

[0080] Example 2

[0081] Referring to Table 1, which is a second embodiment of the present invention, this embodiment provides a fault analysis method for a power transmission and transformation control system. In order to verify the beneficial effects of the present invention, scientific demonstration is carried out through economic benefit calculation and simulation experiments.

[0082] Specifically, the experiment was conducted in a 330kV substation, and the operating data from October 2023 to March 2024 were selected for analysis. The experimental environment includes: 3 main transformers (#1, #2 and #3 main transformers), 12 circuit breakers, 24 disconnectors and 3 sets of voltage transformers and current transformers. A distributed data acquisition system was used, with a sampling frequency of 100Hz and a sampling period of 10ms. Three types of data were collected: (1) primary equipment data, including phase voltage, current, active power and reactive power; (2) secondary equipment data, including protection device action signals, equipment status signals and alarm signals; (3) environmental monitoring data, including equipment surface temperature, ambient humidity and equipment vibration values. The collected data were timestamped in units of 10ms to construct a 6000×21 time series data matrix.

[0083] Furthermore, an improved variational modal decomposition algorithm was used, with an initial mode number K = 8, a bandwidth limit τ = 0.1, and a maximum number of iterations R = 500. A quadratic programming optimization algorithm was used to solve the problem, stopping iterations when the bandwidth change was less than 0.001. The obtained natural modal components were sorted from high to low by center frequency, and the main components carrying fault characteristics were screened.

[0084] Furthermore, during the feature extraction phase, 12 time-domain features (mean, variance, and peak factor) and 8 frequency-domain features (frequency center and power spectrum entropy) were calculated to construct a 20-dimensional feature space. An improved density peak clustering algorithm was used, with the local density calculation radius dc set to 2% of the total number of samples and the cutoff distance set to 10% of the maximum distance in the feature space. In fault correlation analysis, a similarity threshold θ was set to 0.75, and an improved cosine similarity algorithm that accounts for spatiotemporal attenuation was used. In node importance calculation, the degree centrality weight α was 0.4, the betweenness centrality weight β was 0.3, and the similarity weight γ was 0.3.

[0085] Furthermore, during the fault diagnosis phase, a handling strategy library containing 200 typical fault cases was established. In the solution matching calculation, the device matching weight w1 = 0.35, the location matching weight w2 = 0.30, and the historical success rate weight w3 = 0.35. The handling priority P is divided into four levels, and the response time window T is set to 15 minutes, 2 hours, 8 hours, and 24 hours, depending on the fault severity.

[0086] Specifically, as shown in Table 1, the performance comparison between the method of the present invention and the traditional FFT method in the field of power transmission and transformation control system fault analysis is shown. In terms of signal processing, the present invention uses variational mode decomposition (VMD) technology, while the traditional method uses fast Fourier transform.

[0087] Table 1 Comparison of fault analysis methods for power transmission and transformation control systems

[0088] Comparison Project Method of the present invention Traditional FFT method Signal processing method Variational Mode Decomposition (VMD) Fast Fourier Transform Component adaptability (%) 95.8 75.2 Noise immunity (%) 93.5 72.8 Fault identification rate (%) 96.8 82.5 Modal aliasing degree (%) 5.2 28.6 Feature extraction accuracy (%) 94.2 78.3 Fault location accuracy (%) 95.5 76.8 Treatment plan matching degree (%) 93.8 72.5 System response time (ms) 45 180

[0089] Furthermore, the present invention demonstrates significant advantages in core technical indicators. Component adaptability has been improved from the traditional 75.2% to 95.8%, and noise immunity has been increased from 72.8% to 93.5%. These two improvements directly enhance the system's ability to process complex signals. In particular, the present invention achieves a fault recognition rate of 96.8%, a significant improvement over the traditional FFT method's 82.5%, indicating that this method can more accurately capture fault characteristics.

[0090] Furthermore, in terms of signal quality, the modal aliasing of the present invention is only 5.2%, far lower than the 28.6% of the traditional FFT method, demonstrating that the original information is better preserved during signal processing. In terms of feature extraction accuracy and fault location precision, the present invention achieved 94.2% and 95.5%, respectively, compared to the 78.3% and 76.8% of traditional methods, respectively. This significantly improves the accuracy of fault diagnosis.

[0091] Furthermore, the present invention also demonstrates excellent performance in practical applications. The matching rate of processing solutions has increased from the traditional 72.5% to 93.8%, indicating that the system can more accurately select the appropriate fault handling solution. Particularly noteworthy is the reduction in system response time from 180ms to 45ms, a fourfold increase in response speed, which is of great significance in practical engineering applications.

[0092] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A fault analysis method for a power transmission and transformation control system, characterized by: include, Collecting operating data from the power transmission and transformation control system, and grouping and labeling the operating data to form a time series data matrix; Adopting variational mode decomposition method to perform adaptive component extraction on the time series data matrix to obtain intrinsic mode components; Based on the inherent modal components, a feature space including multi-dimensional features is constructed to extract fault feature points; Calculate the similarity between the fault feature point and the historical fault features in the fault sample library to generate a fault correlation topology graph G, wherein the node correlation strength in the fault correlation topology graph G is used to classify the fault degree; Retrieve the corresponding processing solution from the preset fault processing strategy library according to the fault level and output the corresponding fault cause analysis results; Establish a mapping relationship table M between fault levels and processing solutions in a preset fault handling strategy library, wherein the mapping relationship table M includes the fault device type, fault location, fault level, processing priority P, and response time window T; According to the node importance index Wi in the fault association topology graph G, a set of processing solutions S corresponding to the fault level in the mapping relationship table M is retrieved, wherein each solution in the set of processing solutions S includes a description of the fault cause, a sequence of treatment measures, and maintenance resource requirements; Calculate the matching score for each solution in the solution set S , the specific formula is as follows: ； in, is the matching score of fault handling solution i, is the device matching weight factor, is the degree of device matching, is the position matching weight factor, is the position matching degree, is the historical success rate weight factor, is the historical processing success rate, is the impact factor, is the weight coefficient of node u, is the correlation function, is the success probability of the jth processing sub-strategy, is the total number of network nodes, r is the total number of processing sub-strategies; The solution with the highest matching score is selected as the optimal solution, and an analysis report R is generated. The solutions are sorted according to the treatment priority P, and the response time window T of each treatment measure is marked. The analysis report R includes the fault cause chain, key influencing factors, and treatment suggestions. The method for generating the fault correlation topology graph G is: The similarity value between the feature vectors of the fault feature points and the historical fault features in the fault sample library is calculated using the cosine similarity function. The specific formula is as follows: ； in, is the similarity value between the fault feature point a and the historical fault feature point b, is the kth component of the ath eigenvector, is the kth component of the bth eigenvector, m is the feature dimension, is the distance decay function; According to the similarity value Construct the adjacency matrix A, when the similarity value When the similarity value is greater than the similarity threshold θ, the corresponding position in the adjacency matrix A is marked as 1; when the similarity value is greater than the similarity threshold θ, the corresponding position in the adjacency matrix A is marked as 1; When it is less than or equal to the similarity threshold θ, the corresponding position in the adjacency matrix A is marked as 0; The adjacency matrix A is used to generate a fault correlation topology graph G, wherein the nodes of the fault correlation topology graph G represent fault feature points, the edges represent the association relationship between nodes, and the weights of the edges are set to the corresponding similarity values. ; Calculate the degree centrality of the nodes in the fault association topology graph G and betweenness centrality , and combined with the similarity value Calculate node importance index , the specific formula is as follows: ； in, is the importance index of node i, is the degree centrality weight coefficient, is the degree centrality of node i, is the betweenness centrality of node i, is the betweenness centrality weight coefficient, is the similarity weight coefficient, V is the total number of network nodes; According to the node importance index The fault degree is divided into level one fault, level two fault, level three fault and level four fault; The method for extracting the fault feature points is: For each natural mode component Calculate the time domain eigenvalue and frequency domain eigenvalue respectively; The time domain eigenvalue and the frequency domain eigenvalue are combined into a eigenvector F, and normalized; Constructing a Y-dimensional feature space according to the processed feature vector F, wherein each coordinate axis in the Y-dimensional feature space corresponds to a feature component in the feature vector F; Based on the Y-dimensional feature space, data points with similar eigenvalues ​​are clustered to form feature clusters, and the local density and relative distance of each feature cluster are calculated using a density peak clustering algorithm; Based on the product of the local density and the relative distance, the feature cluster center with the maximum product value is selected as the fault feature point.

2. The fault analysis method for a power transmission and transformation control system according to claim 1, wherein: The method for obtaining the intrinsic modal components is: Perform variational mode decomposition on each type of operating data in the time series data matrix, decompose each time series signal into K natural mode components, and calculate the center frequency and bandwidth constraints; Through the secondary optimization iterative calculation, the bandwidth constraint of adjacent natural mode components is minimized. When the number of iterations reaches the preset value R or the bandwidth change is less than the threshold, the iteration is stopped and the natural mode components are sorted from high to low according to the center frequency to form a component sequence. .

3. The fault analysis method for a power transmission and transformation control system according to claim 2, wherein: The method for obtaining the time series data matrix is: Collecting operating data from the power transmission and transformation control system, wherein the operating data includes primary equipment data, secondary equipment data and environmental monitoring data; The operation data are timestamped according to the sampling time points, and the operation data with the same timestamp are combined along the time axis with the sampling period T as the unit; An M×N time series data matrix is ​​established for the annotated operating data, wherein the time series data matrix includes normal operating data and fault operating data, M represents the number of sampling time points, and N represents the number of types of operating data.

4. The fault analysis method for a power transmission and transformation control system according to claim 3, wherein: The primary equipment data includes voltage value, current value, active power value and reactive power value; the secondary equipment data includes protection action signal, alarm signal and switch status signal; the environmental monitoring data includes temperature value, humidity value and vibration value.

5. A fault analysis system for a power transmission and transformation control system, based on the fault analysis method for a power transmission and transformation control system according to any one of claims 1 to 4, characterized in that: include, The collection and annotation module is used to collect the operating data in the power transmission and transformation control system, and group and annotate the operating data to form a time series data matrix; An extraction module, configured to perform adaptive component extraction on the time series data matrix using a variational mode decomposition method to obtain intrinsic mode components; A construction module, based on the inherent modal components, constructs a feature space containing multi-dimensional features and extracts fault feature points; A generation module is used to calculate the similarity between the fault feature points and the historical fault features in the fault sample library to generate a fault correlation topology graph G, wherein the node correlation strength in the fault correlation topology graph G is used to classify the fault degree; The analysis module is used to retrieve the corresponding processing solution from the preset fault processing strategy library according to the fault level and output the corresponding fault cause analysis results.

6. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the fault analysis method for a power transmission and transformation control system according to any one of claims 1 to 4 are implemented.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the fault analysis method for a power transmission and transformation control system according to any one of claims 1 to 4 are implemented.

Images