Relay protection setting value analysis and verification method and system

Abstract

The invention discloses a relay protection setting value analysis and verification method and system, and relates to the technical field of relay protection, and the method comprises the steps: calculating the final coupling strength through employing an exponential coupling formula method, and constructing a coupling matrix through employing a coupling matrix standardization method; performing eigendecomposition on the updated manifold center by using a eigendecomposition verification method, organizing into a third-order tensor, calculating Gaussian kernel similarity between the fault feature and the historical fault feature by using a Gaussian kernel similarity calculation method, and generating a similarity score; and calculating constant value deviation and updating. By introducing differential geometry, information geometry and quantum state coupling theories, a modal covariance tensor model based on a symmetric positive definite manifold is constructed, an index coupling mechanism combining an inter-modal Riemannian distance, a tension index and disturbance coding is put forward, the structural integrity and coupling strength characterization capability of modal information fusion are improved, and the accuracy of modal information fusion is improved. And the scientificity and operability of constant value evaluation are enhanced.

Description

Technical Field

[0001] The present invention relates to the technical field of relay protection, and in particular to a relay protection setting value analysis and verification method and system. Background Art

[0002] With the development of smart grids and automated power systems, relay protection technology, as one of the key technologies to ensure the safe and stable operation of the power grid, has a vital impact on the operation of the power system. The rationality and accuracy of its setting values ​​have a vital impact on the operation of the power system. Traditional relay protection setting values ​​are mainly calculated manually based on electrical parameters such as short-circuit current and load changes, and verified by simulation analysis or empirical judgment. With the rapid development of information and communication technology, a large amount of operation, maintenance and environmental data are collected in real time, providing a rich data foundation for intelligent analysis and auxiliary verification of relay protection setting values.

[0003] The existing relay protection setting verification methods still have shortcomings. The structural heterogeneity and temporal inconsistency of multimodal data make it impossible for traditional data fusion methods to accurately express the coupling mechanism between modes, resulting in error accumulation and uncertainty in the analysis results. Summary of the Invention

[0004] In view of the above existing problems, the present invention is proposed.

[0005] Therefore, the present invention provides a method and system for analyzing and verifying relay protection settings to solve the problem that the structural heterogeneity and temporal inconsistency of multimodal data make it impossible for traditional data fusion methods to accurately express the coupling mechanism between modes, resulting in error accumulation and uncertainty in the analysis results.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0007] In a first aspect, the present invention provides a method for analyzing and verifying relay protection settings, which comprises:

[0008] Three types of modal data were collected and preprocessed. The covariance matrix of each mode was calculated using the modal covariance matrix construction method. The covariance matrix of each mode was embedded into a symmetric positive definite manifold using the symmetric positive definite manifold embedding method, and the covariance matrix was regarded as a point on the symmetric positive definite manifold. The Riemann distance between the modes was calculated, and the tension index was calculated using the matrix trace calculation method. Based on the Riemann distance and tension index, the exponential coupling formula method was used to calculate the final coupling strength. The coupling matrix was constructed using the coupling matrix normalization method.

[0009] The manifold center of all modal covariance matrices is calculated using the Karcher average iteration method. The updated manifold center is decomposed using the eigendecomposition verification method and organized into a third-order tensor. The modal slices of the third-order tensor are normalized and weighted using a coupling matrix to generate a fused feature matrix. The L1 regularized sparse coding method is used to construct an optimization problem. A support vector machine model is built to predict fault detection results. The Gaussian kernel similarity calculation method is used to calculate the Gaussian kernel similarity between the fault signature and the historical fault signature to generate a similarity score.

[0010] Linear projection is used to map fault features to the constant value space, fault feature vectors are generated, constant value deviations and updates are calculated, and a visualization interface is constructed to display the three types of modal data.

[0011] As a preferred solution of the relay protection setting value analysis and verification method of the present invention, wherein: the final coupling strength is calculated using the exponential coupling formula method, and the coupling matrix is ​​constructed using the coupling matrix normalization method, including:

[0012] The multimodal feature matrix construction method is used to splice the three types of modal data into feature vectors according to the time window, and the feature vectors of the modalities are stacked row by row to generate the modal feature matrix;

[0013] Based on the modal feature matrix, the phase time and mean vector of each column feature are calculated, and the phase time is organized into a matrix according to the time window and feature dimension to obtain the priority matrix;

[0014] The modal covariance matrix construction method is used to calculate the covariance matrix of each mode, and the singular value decomposition method is used to perform eigendecomposition on the covariance matrix to obtain the eigendecomposition result;

[0015] The covariance matrix of each mode is embedded into a symmetric positive definite manifold using the symmetric positive definite manifold embedding method, and is regarded as a point on the symmetric positive definite manifold. Based on the eigendecomposition results, the Riemann distance between the modes is calculated, and the tension index is calculated using the matrix trace calculation method.

[0016] The perturbation sequence generation method is used to calculate the perturbation sequence matrix. The perturbation coding matrix is ​​generated based on the element-by-element addition of the priority matrix and the perturbation sequence. The frequency domain inner product projection method is used to calculate the modulation coupling strength between modes. Based on the Riemann distance and tension index, the exponential coupling formula method is used to calculate the final coupling strength. The coupling matrix is ​​constructed using the coupling matrix normalization method.

[0017] As a preferred solution of the relay protection setting value analysis and verification method of the present invention, the updated manifold center is subjected to characteristic decomposition using the characteristic decomposition verification method and organized into a third-order tensor, including:

[0018] The manifold center of all modal covariance matrices is calculated using the Karcher average iteration method. The maximum number of iterations is set using the convergence threshold judgment method. When the maximum number of iterations is reached, the iteration is stopped to obtain the updated manifold center. The updated manifold center is decomposed using the eigendecomposition verification method. The first r principal components are extracted using the principal component extraction method to generate a fused eigenvector.

[0019] The fused feature vectors are organized into a third-order tensor using the tensor organization method.

[0020] As a preferred solution of the relay protection setting value analysis and verification method of the present invention, wherein: the Gaussian kernel similarity calculation method is used to calculate the Gaussian kernel similarity between the fault characteristics and the historical fault characteristics to generate a similarity score, including:

[0021] A tensor slice extraction method is used to extract modal slices from a third-order tensor. The modal slices of the third-order tensor are normalized and weighted using a coupling matrix to generate a fusion feature matrix. A dictionary matrix is ​​randomly initialized. Based on the fusion feature matrix and the dictionary matrix, an optimization problem is constructed using an L1 regularized sparse coding method. The problem is solved using the proximal gradient method to obtain a sparse coefficient matrix. The sparse coefficient matrix is ​​transposed to generate a constant-value check feature.

[0022] Based on the dictionary matrix and the sparse coefficient matrix, matrix multiplication is used to calculate the reconstructed feature matrix and the reconstruction error. The error threshold is set using the empirical threshold setting method. If the reconstruction error is less than the error threshold, the sparse coding is considered to be effective. Otherwise, the sparse regularization parameters are adjusted using the parameter iterative adjustment method until the reconstruction error is less than the error threshold.

[0023] A support vector machine model is constructed to predict fault detection results. Based on the fault detection results, fault features are screened from the fixed value verification features and grouped according to fault types to form fault features.

[0024] Historical fault features are collected from the historical fault scenario database. The Gaussian kernel similarity calculation method is used to calculate the Gaussian kernel similarity between the fault features and the historical fault features, and normalized processing is performed to generate a similarity score.

[0025] As a preferred solution of the relay protection setting value analysis and verification method of the present invention, the calculation and update of the setting value deviation includes:

[0026] The original fixed value parameters, including voltage fixed value, current fixed value and power fixed value, are extracted from the relay protection system configuration of the substation using the fixed value initialization extraction method, and the original fixed value feature vector is generated using the direct splicing method.

[0027] The fault feature with the highest similarity score is selected and mapped to the fixed-value space using linear projection to generate a fault feature vector. Based on the original fixed-value feature vector, the fixed-value deviation is calculated and the fixed-value deviation is used to update the original fixed-value vector to obtain the corrected fixed-value vector.

[0028] As a preferred solution of the relay protection setting value analysis and verification method of the present invention, the construction of a visual interface displays three types of modal data, including:

[0029] A visualization interface is constructed to visualize the three types of modal data collected and the corrected constant vectors, allowing users who have passed real-name verification to view them.

[0030] As a preferred solution of the relay protection setting value analysis and verification method of the present invention, the collecting and preprocessing of three types of modal data includes:

[0031] The multimodal data acquisition method is used to obtain three types of modal data from the relay protection system and perform denoising and normalization processing;

[0032] The three types of modal data include three-phase current, three-phase voltage, power factor, operating time, number of maintenance times, number of faults, temperature, humidity, air pressure and wind speed data.

[0033] In a second aspect, the present invention provides a relay protection setting value analysis and verification system, comprising:

[0034] The collection and construction module is used to collect three types of modal data and perform preprocessing. The modal covariance matrix construction method is used to calculate the covariance matrix of each mode. The symmetric positive definite manifold embedding method is used to embed the covariance matrix of each mode into a symmetric positive definite manifold, treating it as a point on the symmetric positive definite manifold. The Riemann distance between the modes is calculated. The matrix trace calculation method is used to calculate the tension index. Based on the Riemann distance and tension index, the exponential coupling formula method is used to calculate the final coupling strength. The coupling matrix is ​​constructed using the coupling matrix normalization method.

[0035] The tensor score module is used to calculate the manifold center of all modal covariance matrices using the Karcher average iteration method, perform eigendecomposition on the updated manifold center using the eigendecomposition verification method, organize it into a third-order tensor, normalize and weight the modal slices of the third-order tensor using the coupling matrix, generate a fused feature matrix, construct the optimization problem using the L1 regularized sparse coding method, build a support vector machine model to predict fault detection results, and use the Gaussian kernel similarity calculation method to calculate the Gaussian kernel similarity between the fault signature and the historical fault signature to generate a similarity score;

[0036] The verification visualization module is used to map fault features to the constant value space using linear projection, generate fault feature vectors, calculate constant value deviations and updates, and build a visualization interface to display the three types of modal data.

[0037] In a third aspect, 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 is executed by the processor, any step of the relay protection constant value analysis and verification method described in the first aspect of the present invention is implemented.

[0038] In a fourth aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, any step of the relay protection constant value analysis and verification method as described in the first aspect of the present invention is implemented.

[0039] The beneficial effects of the present invention are as follows: the present invention introduces differential geometry, information geometry and quantum state coupling theory, constructs a modal covariance tensor model based on symmetric positive definite manifolds, proposes an exponential coupling mechanism combining the Riemann distance between modes, tension index and disturbance coding, improves the structural integrity and coupling strength characterization capability of modal information fusion, and enhances the scientificity and operability of constant value evaluation. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0041] Figure 1 This is a flow chart of the relay protection constant analysis and verification method in Example 1.

[0042] Figure 2 This is a schematic diagram of the relay protection constant value analysis and verification system in Example 1.

[0043] Figure 3 Schematic diagram of the embedded symmetric positive definite manifold in the relay protection constant value analysis and verification method in Example 1.

[0044] Figure 4 This is a schematic diagram showing three types of modal data in a visual interface in the relay protection constant analysis and verification method in Example 1. DETAILED DESCRIPTION

[0045] 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.

[0046] 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.

[0047] 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.

[0048] Example 1, with reference to Figures 1 to 4 , which is the first embodiment of the present invention, provides a relay protection setting value analysis and verification method, comprising the following steps:

[0049] S1. Collect three types of modal data and perform preprocessing. Use the modal covariance matrix construction method to calculate the covariance matrix of each mode. Use the symmetric positive definite manifold embedding method to embed the covariance matrix of each mode into a symmetric positive definite manifold, treating it as a point on the symmetric positive definite manifold. Calculate the Riemann distance between the modes. Use the matrix trace calculation method to calculate the tension index. Based on the Riemann distance and tension index, use the exponential coupling formula method to calculate the final coupling strength. Use the coupling matrix normalization method to construct the coupling matrix.

[0050] Specifically, three types of modal data are collected and preprocessed, including:

[0051] The multimodal data acquisition method is used to obtain three types of modal data from the relay protection system and perform denoising and normalization processing;

[0052] The three types of modal data include three-phase current, three-phase voltage, power factor, operating time, number of maintenance times, number of faults, temperature, humidity, air pressure and wind speed data.

[0053] Multimodal acquisition enhances the all-round perception capability of the system status. The preprocessing process effectively improves the accuracy and robustness of subsequent feature analysis. Multimodal input enhances the redundancy and information diversity of the system, providing a solid data foundation for complex state modeling.

[0054] Furthermore, the final coupling strength is calculated using the exponential coupling formula method, and the coupling matrix is ​​constructed using the coupling matrix normalization method, including:

[0055] The multimodal feature matrix construction method is used to splice the three types of modal data into feature vectors according to the time window (set using the fixed time window method). The modal feature vectors are stacked row by row to generate the modal feature matrix (including electrical mode (three-phase current, three-phase voltage and power factor), operation and maintenance mode (operation time, number of maintenance and number of faults) and environmental mode matrix (temperature, humidity, air pressure and wind speed)).

[0056] Based on the feature matrix of the modality, the phase time and mean vector of each column feature are calculated, and the phase time is organized into a matrix according to the time window and feature dimension to obtain the priority matrix. The rows represent samples and the columns represent feature dimensions. The formula is:

[0057]

[0058] Among them, P i [j,k] is the phase time of mode i in the jth time window and the kth feature, X i [j,k] is the kth eigenvalue of mode i in the jth time window, rank(X i [j,k]in X i [:,k]) is X i [j,k] is the kth column of X i rank in [:,k], is the feature mean vector of mode i, n is the number of time windows, is the eigenvector of the jth time window of mode i;

[0059] The modal covariance matrix construction method is used to calculate the covariance matrix of each mode. The formula is:

[0060]

[0061] where Σ i is the covariance matrix of mode i, T is the transpose;

[0062] Use the singular value decomposition method to perform eigendecomposition on the covariance matrix to obtain the eigendecomposition results, including the eigenvector matrix and the eigenvalue diagonal matrix;

[0063] The symmetric positive definite manifold embedding method is used to embed the covariance matrix of each mode into a symmetric positive definite manifold and regard it as a point on the symmetric positive definite manifold. Based on the eigendecomposition results, the Riemann distance between the modes is calculated as follows:

[0064]

[0065] Among them D R (Q i ,Q u ) is the Riemann distance between mode i and mode u, Qi and Q u are the covariance matrices of modes i and u, respectively, embedded in a point on the symmetric positive definite manifold, λ k is the kth eigenvalue of the similarity transformation matrix, and d is the minimum common characteristic dimension of the similarity transformation matrix;

[0066] The inverse square root of the covariance matrix is ​​calculated using the inverse square root method, as follows:

[0067]

[0068] in is the inverse square root of the covariance matrix of mode i, U i and Λ i are the eigendecomposition results of mode i respectively;

[0069] The similarity transformation matrix is ​​calculated using the matrix similarity transformation method. The formula is:

[0070]

[0071] Where A is the similarity transformation matrix, Σ u is the covariance matrix of mode u;

[0072] Use the eigenvalue extraction method to perform eigenvalue decomposition on the similarity transformation matrix to obtain the eigenvalue of the similarity transformation matrix;

[0073] The tension index is calculated using the matrix trace method, and the formula is:

[0074]

[0075] Among them O iu is the tension index between mode i and mode u, tr(·) is the matrix trace, the sum of the main diagonal elements;

[0076] The perturbation sequence generation method is used to calculate the perturbation sequence matrix. The formula is:

[0077]

[0078] where Δ i is the perturbation sequence matrix of mode i, the rows represent samples, the columns represent feature dimensions, d i is the characteristic dimension of mode i, σ i,k For X i The standard deviation of [:,k], φ i is the phase offset of mode i (1 represents the electrical mode, 2 represents the operation and maintenance mode, and 3 represents the environmental mode, which are set using the fixed phase offset method);

[0079] Based on the element-by-element addition of the priority matrix and the perturbation sequence, the perturbation coding matrix is ​​generated, and the formula is:

[0080]

[0081] Among them, M i is the perturbation encoding matrix of mode i, P i is the priority matrix of mode i;

[0082] The modulation coupling strength between modes is calculated using the frequency domain inner product projection method, and the formula is:

[0083]

[0084] in is the modulation coupling strength of modes i and u, M i [j,k] and M u [j,k] are the perturbation encoding values ​​of modal i and u in the jth time window and the kth feature respectively, and is the self-modulation coupling strength of modes i and u;

[0085] Based on the Riemann distance and tension index, the exponential coupling formula method is used to calculate the final coupling strength, and the coupling matrix is ​​constructed using the coupling matrix normalization method. The formula is:

[0086]

[0087] Among them C iu is the final coupling strength between mode i and mode u, α is the fusion weight, which controls the ratio of geometric and frequency domain contributions (set using the fixed weight assignment method), C ii and C uu are the final coupling strengths of modes i and u, respectively.

[0088] The modes are represented by the covariance matrix, and the statistical characteristic space of the modes is constructed to achieve the mathematical reduction of the modes. Symmetric positive manifolds are used to improve the geometric accuracy of the modal expression, ensure that the similarity calculation reflects the statistical structure differences, avoid the errors caused by coordinate rotation or scale mismatch in the modal characteristics in the Euclidean space, and improve the robustness and credibility of the similarity calculation between modes. The introduction of matrix trace as a representation of the modal coupling strength is an effective way to combine the frequency domain response characteristics. The sum of its main diagonal elements reflects the strength of the main components of the modal coupling, avoids the use of high-dimensional frequency domain spectra for direct comparison, and improves the computational efficiency. Through the use of traces, the main component response of the modal coupling is focused, and non-main disturbances are ignored, which effectively improves the accuracy of the modulation intensity estimation. The disturbance sequence coding is introduced into the modal coupling analysis. On the basis of traditional modal comparison The disturbance response sensitivity analysis is introduced to simulate the coupling reaction of the mode to noise or external influences, enhance the coupling analysis capability of the system in a dynamic environment, improve the actual adaptability of the modeling, and reflect the changing response of different modes in real complex scenarios through disturbance simulation, thereby improving the dynamic adaptability of the coupling strength assessment. Traditional modal coupling methods are mostly based on a single distance or correlation index. The present invention integrates the Riemann distance and the tension index into the exponential coupling formula to construct a multi-level coupling strength modeling. The exponential function has a nonlinear amplification characteristic, which makes the strong coupling relationship more prominent at the exponential scale, and the weak relationship approaches zero, which is convenient for subsequent clustering or discrimination, and effectively solves the problem of the numerical distribution of the coupling degree between modes being too concentrated or too discrete, so that the coupling matrix has good distribution characteristics, which is convenient for structural analysis.

[0089] S2. Use the Karcher average iteration method to calculate the manifold center of all modal covariance matrices, use the eigendecomposition verification method to perform eigendecomposition on the updated manifold center and organize it into a third-order tensor. Use the coupling matrix to normalize and weight the modal slices of the third-order tensor to generate a fused feature matrix. Use the L1 regularized sparse coding method to construct the optimization problem, build a support vector machine model to predict fault detection results, and use the Gaussian kernel similarity calculation method to calculate the Gaussian kernel similarity between the fault characteristics and historical fault characteristics to generate a similarity score.

[0090] Specifically, the updated manifold center is decomposed using the eigendecomposition verification method and organized into a third-order tensor, including:

[0091] The Karcher average iteration method is used to calculate the manifold center of all modal covariance matrices. The maximum number of iterations is set using the convergence threshold judgment method. When the maximum number of iterations is reached, the iteration is stopped and the updated manifold center is obtained. The formula is:

[0092]

[0093] where Σ cis the manifold center covariance matrix, the geometric center of the three modal covariance matrices on the symmetric positive definite manifold, is the square of the Riemann distance between the covariance matrix of mode i and the symmetric positive definite matrix Σ, Σ (t) is the manifold center of the tth iteration;

[0094] Use the eigendecomposition verification method to perform eigendecomposition on the updated manifold center, use the principal component extraction method to extract the first r principal components, where r is the number of principal components, and generate a fusion feature vector;

[0095] The fused feature vectors are organized into a third-order tensor using the tensor organization method.

[0096] The Karcher average iteration method utilizes the geometric properties of Riemannian manifolds and captures the statistical distribution characteristics of multimodal data by iteratively approximating the center of the manifold, solving the mismatch problem of traditional mean methods on high-dimensional nonlinear data. The application of Riemann distance improves the representativeness of the modal center and enhances the accuracy of subsequent feature decomposition. The introduction of third-order tensors organizes multimodal data into a high-dimensional structure, retains the interactive information between modalities, and solves the problems of dimensional inconsistency and information redundancy in multimodal data.

[0097] Furthermore, the Gaussian kernel similarity calculation method is used to calculate the Gaussian kernel similarity between the fault signature and the historical fault signature to generate a similarity score, including:

[0098] The modal slices are extracted from the third-order tensor using the tensor slice extraction method. The modal slices of the third-order tensor are normalized and weighted using the coupling matrix to generate a fusion feature matrix. The formula is:

[0099]

[0100] in is the weighted fusion feature matrix, is a tensor The slice of the i-th mode;

[0101] Randomly initialize the dictionary matrix, and based on the fusion feature matrix and the dictionary matrix, use the L1 regularized sparse coding method to construct the optimization problem. The formula is:

[0102]

[0103] Where A is a sparse coefficient matrix, is the L1 regularization parameter, controlling the sparsity, ‖·‖ F is the Frobenius norm, which calculates the square root of the sum of squares of the matrix;

[0104] Use the proximal gradient method to solve and obtain a sparse coefficient matrix, transpose the sparse coefficient matrix and generate a fixed value verification feature;

[0105] Based on the dictionary matrix and the sparse coefficient matrix, matrix multiplication is used to calculate the reconstructed feature matrix, calculate the reconstruction error, and use the empirical threshold setting method to set the error threshold. If the reconstruction error is less than the error threshold, it is judged that the sparse coding is effective. Otherwise, the parameter iterative adjustment method is used to adjust the sparse regularization parameters until the reconstruction error is less than the error threshold. The adjustment stops. The formula is:

[0106]

[0107] Where E is the reconstruction error, is the reconstructed feature matrix, which represents the features reconstructed by sparse coding;

[0108] Collect historical three-category modal data with labels, calculate fixed value verification features, and generate a training set;

[0109] Build a support vector machine model, train the support vector machine model using the training set, and iteratively optimize the support vector machine model parameters using the loss function and Adam optimizer;

[0110] Input the fixed value verification features into the trained support vector machine model to obtain the fault detection results and the corresponding fault types (such as normal, short circuit and grounding);

[0111] Based on the fault detection results, the fault characteristics are screened out from the fixed value verification characteristics, and grouped according to the fault type to form the fault characteristics;

[0112] Historical fault features are collected from the historical fault scenario database. The Gaussian kernel similarity calculation method is used to calculate the Gaussian kernel similarity between the fault features and the historical fault features, and normalized processing is performed to generate a similarity score.

[0113] Normalized weighting ensures that each mode (such as voltage and current) contributes evenly to the fusion features. The optimization problem is constructed through L1 regularized sparse coding, and the sparse coefficient matrix is ​​solved in combination with the proximal gradient method. The parameters are adaptively adjusted through reconstruction error and empirical threshold, avoiding the model deviation caused by the artificial setting of regularization parameters in traditional sparse coding and enhancing the adaptive ability of feature expression. The Gaussian kernel can capture the complex relationship between features, improve the accuracy of ground fault identification, avoid the deviation problem caused by information imbalance between modes, and improve the structural consistency and semantic complementarity of modal information through the coupling matrix.

[0114] S3. Use linear projection to map the fault characteristics to the fixed value space, generate the fault feature vector, calculate the fixed value deviation and update, and build a visualization interface to display the three types of modal data;

[0115] Specifically, calculation of fixed value deviation and update includes:

[0116] The original fixed value parameters, including voltage fixed value, current fixed value and power fixed value, are extracted from the relay protection system configuration of the substation using the fixed value initialization extraction method, and the original fixed value feature vector is generated using the direct splicing method.

[0117] Select the fault feature with the highest similarity score, use linear projection to map the fault feature to the fixed value space, generate the fault feature vector, and calculate the fixed value deviation based on the original fixed value feature vector. The formula is:

[0118]

[0119] Where ΔV is the fixed value deviation vector, is the learning rate (set using the empirical learning rate setting method), is the fault feature vector, V set is the original fixed-valued eigenvector;

[0120] Use the fixed value deviation to update the original fixed value vector to obtain the corrected fixed value vector. The formula is:

[0121] V new =V set +ΔV,

[0122] Where V new is the corrected constant vector.

[0123] The direct splicing method integrates multi-dimensional constant parameters into a single vector, simplifying the complexity of subsequent calculations. Linear projection can reduce high-dimensional and complex fault characteristics to the same space as the constant parameters, retaining key information while reducing computational complexity. The adjustment direction and amplitude are clearly quantified through vector difference. The introduction of learning rate avoids false triggering of protection logic due to excessive adjustment. The incremental update strategy ensures that the constant adjustment is based on actual fault characteristics. Compared with fixed rule adjustment, it is more in line with real-time working conditions. The update step size is controlled by learning rate, which avoids false operation or refusal of protection system due to excessive adjustment.

[0124] Furthermore, a visualization interface is constructed to display three types of modal data, including:

[0125] A visualization interface is constructed to visualize the three types of modal data collected and the corrected constant vectors, allowing users who have passed real-name verification to view them.

[0126] The visualization interface can display modal data and constant value change trends through charts (such as line charts and bar charts), helping operation and maintenance personnel to quickly identify anomalies. The real-name verification mechanism ensures that only authorized users can access sensitive data, complying with power system information security standards.

[0127] This embodiment also provides a relay protection setting value analysis and verification system, including:

[0128] The collection and construction module is used to collect three types of modal data and perform preprocessing. The modal covariance matrix construction method is used to calculate the covariance matrix of each mode. The symmetric positive definite manifold embedding method is used to embed the covariance matrix of each mode into a symmetric positive definite manifold, treating it as a point on the symmetric positive definite manifold. The Riemann distance between the modes is calculated. The matrix trace calculation method is used to calculate the tension index. Based on the Riemann distance and tension index, the exponential coupling formula method is used to calculate the final coupling strength. The coupling matrix is ​​constructed using the coupling matrix normalization method.

[0129] The tensor score module is used to calculate the manifold center of all modal covariance matrices using the Karcher average iteration method, perform eigendecomposition on the updated manifold center using the eigendecomposition verification method, organize it into a third-order tensor, normalize and weight the modal slices of the third-order tensor using the coupling matrix, generate a fused feature matrix, construct the optimization problem using the L1 regularized sparse coding method, build a support vector machine model to predict fault detection results, and use the Gaussian kernel similarity calculation method to calculate the Gaussian kernel similarity between the fault signature and the historical fault signature to generate a similarity score;

[0130] The verification visualization module is used to map fault features to the constant value space using linear projection, generate fault feature vectors, calculate constant value deviations and updates, and build a visualization interface to display the three types of modal data.

[0131] This embodiment also provides a computer device, which is suitable for the relay protection constant value analysis and verification method, including: 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 relay protection constant value analysis and verification method proposed in the above embodiment.

[0132] 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.

[0133] This embodiment also provides a storage medium on which a computer program is stored. When the program is executed by a processor, the method for analyzing and verifying the relay protection setting value proposed in the above embodiment is implemented; the storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (Static Random Access Memory, abbreviated as SRAM), electrically erasable programmable read-only memory (Electrically Erasable Programmable Read-Only Memory, abbreviated as EEPROM), erasable programmable read-only memory (Erasable Programmable Read Only Memory, abbreviated as EPROM), programmable read-only memory (Programmable Red-Only Memory, abbreviated as PROM), read-only memory (Read-Only Memory, abbreviated as ROM), magnetic memory, flash memory, disk or optical disk.

[0134] In summary, the present invention introduces differential geometry, information geometry and quantum state coupling theory to construct a modal covariance tensor model based on symmetric positive definite manifolds, and proposes an exponential coupling mechanism combining the Riemann distance between modes, tension index and perturbation coding, so as to improve the structural integrity and coupling strength characterization ability of modal information fusion, and enhance the scientificity and operability of constant value evaluation.

[0135] 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 relay protection setting value analysis and verification method, characterized by: include, Three types of modal data were collected and preprocessed. The covariance matrix of each mode was calculated using the modal covariance matrix construction method. The covariance matrix of each mode was embedded into a symmetric positive definite manifold using the symmetric positive definite manifold embedding method, and the covariance matrix was regarded as a point on the symmetric positive definite manifold. The Riemann distance between the modes was calculated, and the tension index was calculated using the matrix trace calculation method. Based on the Riemann distance and tension index, the exponential coupling formula method was used to calculate the final coupling strength. The coupling matrix was constructed using the coupling matrix normalization method. The manifold center of all modal covariance matrices is calculated using the Karcher average iteration method. The updated manifold center is decomposed using the eigendecomposition verification method and organized into a third-order tensor. The modal slices of the third-order tensor are normalized and weighted using a coupling matrix to generate a fused feature matrix. The L1 regularized sparse coding method is used to construct an optimization problem. A support vector machine model is built to predict fault detection results. The Gaussian kernel similarity calculation method is used to calculate the Gaussian kernel similarity between the fault signature and the historical fault signature to generate a similarity score. Linear projection is used to map fault features to the constant value space, fault feature vectors are generated, constant value deviations and updates are calculated, and a visualization interface is constructed to display the three types of modal data.

2. The relay protection setting value analysis and verification method according to claim 1, characterized in that: The method of calculating the final coupling strength using the exponential coupling formula and constructing the coupling matrix using the coupling matrix normalization method includes: The multimodal feature matrix construction method is used to splice the three types of modal data into feature vectors according to the time window, and the feature vectors of the modalities are stacked row by row to generate the modal feature matrix; Based on the modal feature matrix, the phase time and mean vector of each column feature are calculated, and the phase time is organized into a matrix according to the time window and feature dimension to obtain the priority matrix; The modal covariance matrix construction method is used to calculate the covariance matrix of each mode, and the singular value decomposition method is used to perform eigendecomposition on the covariance matrix to obtain the eigendecomposition result; The covariance matrix of each mode is embedded into a symmetric positive definite manifold using the symmetric positive definite manifold embedding method, and is regarded as a point on the symmetric positive definite manifold. Based on the eigendecomposition results, the Riemann distance between the modes is calculated, and the tension index is calculated using the matrix trace calculation method. The perturbation sequence generation method is used to calculate the perturbation sequence matrix. The perturbation coding matrix is ​​generated based on the element-by-element addition of the priority matrix and the perturbation sequence. The frequency domain inner product projection method is used to calculate the modulation coupling strength between modes. Based on the Riemann distance and tension index, the exponential coupling formula method is used to calculate the final coupling strength. The coupling matrix is ​​constructed using the coupling matrix normalization method.

3. The relay protection setting value analysis and verification method according to claim 2, characterized in that: The updated manifold center is decomposed using the eigendecomposition verification method and organized into a third-order tensor, including: The manifold center of all modal covariance matrices is calculated using the Karcher average iteration method. The maximum number of iterations is set using the convergence threshold judgment method. When the maximum number of iterations is reached, the iteration is stopped to obtain the updated manifold center. The updated manifold center is decomposed using the eigendecomposition verification method. The first r principal components are extracted using the principal component extraction method to generate a fused eigenvector. The fused feature vectors are organized into a third-order tensor using the tensor organization method.

4. The relay protection setting value analysis and verification method according to claim 3, characterized in that: The method of using the Gaussian kernel similarity calculation method to calculate the Gaussian kernel similarity between the fault feature and the historical fault feature to generate a similarity score includes: A tensor slice extraction method is used to extract modal slices from a third-order tensor. The modal slices of the third-order tensor are normalized and weighted using a coupling matrix to generate a fusion feature matrix. A dictionary matrix is ​​randomly initialized. Based on the fusion feature matrix and the dictionary matrix, an optimization problem is constructed using an L1 regularized sparse coding method. The problem is solved using the proximal gradient method to obtain a sparse coefficient matrix. The sparse coefficient matrix is ​​transposed to generate a constant-value check feature. Based on the dictionary matrix and the sparse coefficient matrix, matrix multiplication is used to calculate the reconstructed feature matrix and the reconstruction error. The error threshold is set using the empirical threshold setting method. If the reconstruction error is less than the error threshold, the sparse coding is considered to be effective. Otherwise, the sparse regularization parameters are adjusted using the parameter iterative adjustment method until the reconstruction error is less than the error threshold. A support vector machine model is constructed to predict fault detection results. Based on the fault detection results, fault features are screened from the fixed value verification features and grouped according to fault types to form fault features. Historical fault features are collected from the historical fault scenario database. The Gaussian kernel similarity calculation method is used to calculate the Gaussian kernel similarity between the fault features and the historical fault features, and normalized processing is performed to generate a similarity score.

5. The relay protection setting value analysis and verification method according to claim 4, characterized in that: The calculation and update of the fixed value deviation includes: The original fixed value parameters, including voltage fixed value, current fixed value and power fixed value, are extracted from the relay protection system configuration of the substation using the fixed value initialization extraction method, and the original fixed value feature vector is generated using the direct splicing method. The fault feature with the highest similarity score is selected and mapped to the fixed-value space using linear projection to generate a fault feature vector. Based on the original fixed-value feature vector, the fixed-value deviation is calculated and the fixed-value deviation is used to update the original fixed-value vector to obtain the corrected fixed-value vector.

6. The relay protection setting value analysis and verification method according to claim 5, characterized in that: The visualization interface is constructed to display three types of modal data, including: A visualization interface is constructed to visualize the three types of modal data collected and the corrected constant vectors, allowing users who have passed real-name verification to view them.

7. The relay protection setting value analysis and verification method according to claim 1, characterized in that: The three types of modal data are collected and preprocessed, including: The multimodal data acquisition method is used to obtain three types of modal data from the relay protection system and perform denoising and normalization processing; The three types of modal data include three-phase current, three-phase voltage, power factor, operating time, number of maintenance times, number of faults, temperature, humidity, air pressure and wind speed data.

8. A relay protection setting value analysis and verification system, based on the relay protection setting value analysis and verification method according to any one of claims 1 to 7, characterized in that: include, The collection and construction module is used to collect three types of modal data and perform preprocessing. The modal covariance matrix construction method is used to calculate the covariance matrix of each mode. The symmetric positive definite manifold embedding method is used to embed the covariance matrix of each mode into a symmetric positive definite manifold, treating it as a point on the symmetric positive definite manifold. The Riemann distance between the modes is calculated. The matrix trace calculation method is used to calculate the tension index. Based on the Riemann distance and tension index, the exponential coupling formula method is used to calculate the final coupling strength. The coupling matrix is ​​constructed using the coupling matrix normalization method. The tensor score module is used to calculate the manifold center of all modal covariance matrices using the Karcher average iteration method, perform eigendecomposition on the updated manifold center using the eigendecomposition verification method, organize it into a third-order tensor, normalize and weight the modal slices of the third-order tensor using the coupling matrix, generate a fused feature matrix, construct the optimization problem using the L1 regularized sparse coding method, build a support vector machine model to predict fault detection results, and use the Gaussian kernel similarity calculation method to calculate the Gaussian kernel similarity between the fault signature and the historical fault signature to generate a similarity score; The verification visualization module is used to map fault features to the constant value space using linear projection, generate fault feature vectors, calculate constant value deviations and updates, and build a visualization interface to display the three types of modal data.

9. 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 relay protection setting analysis and verification method described in any one of claims 1 to 7 are implemented.

10. 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 relay protection setting analysis and verification method according to any one of claims 1 to 7 are implemented.

Images