A power system fault severity assessment method based on optical CT measurement

Abstract

The application discloses a kind of power system fault severity assessment methods based on optical CT measurement, it is related to power system technical field, comprising the following steps: S1, construct optical CT measurement model;S2, multiscale decomposition and feature extraction;S3, space-time coupling fault location;S4, dynamic network topology reconstruction;S5, fault influence propagation analysis;S6, power element health state prediction;S7, fault severity quantitative index establishment.The power system fault severity assessment method based on optical CT measurement, by introducing wavelet transform, multiscale decomposition, space-time coupling algorithm and comprehensive severity quantitative index system, the accuracy of fault location and the accuracy of severity assessment are greatly improved, overcome the limitation of traditional technology in fault signal capture and propagation path analysis, can track the dynamic change of power system fault in real time, comprehensively.

Description

Technical Field

[0001] The present invention relates to the technical field of power systems, and in particular to a method for evaluating the severity of power system faults based on optical CT measurement. Background Art

[0002] With the increasing complexity of modern power systems and the advancement of high-voltage transmission and distribution technologies, fault detection and severity assessment have become critical issues in ensuring stable grid operation. Traditional power system fault detection methods, such as those based on current transformers and voltage transformers, are often limited by their inherent physical characteristics and cannot provide real-time, accurate fault location and severity analysis. Especially when multiple fault types (such as instantaneous overcurrent, sustained overload, and line grounding) coexist, traditional methods struggle to fully capture fault characteristics, resulting in delayed response strategies and reduced system stability.

[0003] Power system fault assessment methods based on optical computed tomography (OCT) measurements offer new possibilities for power system fault analysis due to their non-invasive, real-time, and high-precision measurement capabilities. By constructing an OCT measurement model that combines complex electromagnetic field distribution information with the dynamic characteristics of the power system, it is possible to accurately capture internal power system faults. However, existing technologies applying OCT measurement technology are often limited to data acquisition and basic signal processing, lacking in-depth exploration of multi-scale decomposition, spatiotemporal coupling analysis, and severity quantification indicators.

[0004] Existing technologies often rely on fixed-frequency Fourier transforms for multiscale decomposition and feature extraction, making them difficult to adapt to the complex variations of fault signals in the time and frequency domains. Furthermore, traditional fault location methods lack a comprehensive consideration of the spatiotemporal dynamics of power systems, resulting in inaccurate fault propagation path analysis.

[0005] In terms of fault severity assessment, previous quantitative models mostly focused on the analysis of a single parameter and lacked the integration and optimization of multi-dimensional data (such as the amplitude of electromagnetic signal changes, the health status of power components, the speed of fault propagation, etc.). Summary of the Invention

[0006] In view of the deficiencies of the prior art, the present invention provides a method for evaluating the severity of power system faults based on optical CT measurement to solve the problems raised in the above background technology.

[0007] To achieve the above objectives, the present invention provides the following technical solution: a method for evaluating the severity of power system faults based on optical CT measurement, comprising the following steps:

[0008] S1. Constructing optical CT measurement model;

[0009] By establishing an optical CT measurement model to obtain the electromagnetic field distribution in the power system, optical sensors are used to measure the spatial distribution characteristics of electromagnetic signals in real time without affecting the normal operation of the power grid. By installing optical CT sensor arrays at key nodes in the power system, multi-dimensional power signal data is captured. The signal data includes key information such as voltage waveform, current density, and electromagnetic wave propagation path.

[0010] S2, multi-scale decomposition and feature extraction;

[0011] After obtaining the original optical CT measurement signal, multi-scale decomposition is performed to extract the electromagnetic signal characteristics in different frequency bands. Wavelet transform technology is used to decompose the complex power signal into data at multiple scale levels, corresponding to high-frequency and low-frequency components, to capture the unique characteristics of different fault types in the time and frequency domains.

[0012] S3, time-space coupled fault location;

[0013] Based on a spatiotemporal coupling algorithm model, the electromagnetic field data acquired by optical CT is combined with time series to accurately locate the fault point, track the development path of the fault within the power system in real time, and analyze its propagation speed and diffusion range. In this process, the model combines the topological information of the power system to dynamically associate the data of each node with neighboring nodes, identifying the direction and range of possible fault spread.

[0014] S4, dynamic network topology reconstruction;

[0015] After a fault occurs, the network topology model of the power system is updated in real time to reconstruct the relationship between the fault node and its surrounding nodes. Based on the measurement results of optical CT, the electromagnetic coupling relationship within the system is reconfigured to identify the critical paths and components that may be affected by the fault.

[0016] S5. Fault impact propagation analysis;

[0017] Conduct impact propagation analysis on captured electromagnetic signals. By calculating the fault's transmission path and speed in the network, the potential threat to other nodes and regions is assessed, and critical points of fault expansion and high-risk areas are identified, providing a reference for subsequent risk assessment and emergency response.

[0018] S6. Prediction of the health status of power components;

[0019] Use machine learning algorithms to predict the health status of key components in power systems, combined with historical trend analysis of optical CT measurement data to assess the failure probability of components under different fault conditions;

[0020] S7. Establishment of quantitative indicators of fault severity;

[0021] A quantitative indicator system for fault severity is established. Based on the multi-dimensional parameters of the electromagnetic signal change amplitude, power component health status, and fault impact propagation speed obtained by optical CT measurement, the parameters are standardized and assigned different weights to calculate a comprehensive severity score. The scoring system includes different types of fault scenarios, including instantaneous overcurrent, continuous overload, and line grounding, providing a quantitative basis for risk level determination.

[0022] To further optimize the technical solution, in step S1, the constructed optical CT measurement model includes an electromagnetic field distribution equation and a signal reconstruction equation;

[0023] The electromagnetic field distribution equation is as follows:

[0024] ;

[0025] in,

[0026] For the spatial position and time The electromagnetic field distribution on

[0027] is the response function of the optical CT sensor, which depends on the wavelength and sensor azimuth ;

[0028] is the intensity function of the source electromagnetic signal, describing the current or voltage density in the power system over time. and spatial location distribution of

[0029] is the wave vector of the electromagnetic wave, describing the propagation direction;

[0030] is the distance vector between the sensor and the source signal;

[0031] is the angular frequency, which represents the frequency of the electromagnetic wave;

[0032] is the integration area, covering all measurement points in the power system;

[0033] The signal reconstruction equation is as follows:

[0034] ;

[0035] in,

[0036] is the reconstructed three-dimensional electromagnetic field intensity image;

[0037] is the weight coefficient, which represents the contribution of different optical CT sensors in the measurement process;

[0038] For the Sensors at location and time The measurement results on .

[0039] To further optimize this technical solution, the optical CT measurement model includes the following specific processes when used:

[0040] Real-time data acquisition: Utilize optical CT sensor arrays installed at key nodes of the power system to acquire electromagnetic field data in real time. The data is passed through the electromagnetic field distribution equation in the model. Processing is performed to capture the preliminary distribution of fault signals in the power system;

[0041] Fault signal reconstruction: Input the acquired data into the signal reconstruction equation ,Through the weight adjustment mechanism, the three-dimensional distribution image of the electromagnetic field at the time of the fault is reconstructed, and the fault point and the affected area are quickly identified;

[0042] Propagation path analysis: based on wave vector and the distance vector information, track the propagation path of fault signals, and predict their potential impact on other parts of the power system;

[0043] Adaptive response strategy generation: adjust the weight coefficients in the model based on real-time updated optical CT measurement data , generate the optimal fault response strategy, dynamically adjust the load distribution and the action of protection equipment in the power system, and reduce the impact of faults on the system.

[0044] Further optimizing the technical solution, in step S2, a multi-scale decomposition model is constructed based on wavelet transform technology, combined with time-frequency analysis technology to achieve decomposition and feature extraction of signals in different frequency bands;

[0045] The model introduces nonlinear scale adjustment and frequency weighting mechanisms to improve the recognition accuracy and robustness of fault signal features;

[0046] The multi-scale decomposition model includes a multi-scale wavelet decomposition formula, an adaptive frequency weighting function, and a feature extraction formula.

[0047] To further optimize this technical solution, in the multi-scale decomposition model:

[0048] The multi-scale wavelet decomposition formula is as follows:

[0049] ;

[0050] in,

[0051] For scale and frequency The time-frequency coefficient under represents the energy distribution of the signal at different scales and frequency bands;

[0052] is the original optical CT measurement signal;

[0053] For scale and center frequency The mother wavelet function under the control of the fineness of signal decomposition;

[0054] Represents the conjugate complex number of the mother wavelet, which is used to capture the amplitude and phase information of the signal in the time-frequency domain;

[0055] The adaptive frequency weighting function is as follows:

[0056] ;

[0057] in,

[0058] is a frequency weighting function used to adjust the importance of signals in different frequency bands;

[0059] is the scale weight factor, which indicates the signal priority at different scales;

[0060] It is a frequency adjustment parameter used to control the attenuation speed of signals in different frequency bands;

[0061] is the target frequency, representing the fault signal frequency region of interest;

[0062] The feature extraction formula is as follows:

[0063] ;

[0064] in,

[0065] is the final extracted feature vector, representing the comprehensive characteristics of the signal at different scales and frequencies;

[0066] and are discrete series of scale and frequency, respectively;

[0067] It represents the absolute energy value of the signal at a specific scale and frequency.

[0068] To further optimize this technical solution, the multi-scale decomposition model, when used, includes the following specific processes:

[0069] Signal decomposition and preprocessing: After obtaining the original optical CT measurement signal Then, it is input into the multi-scale wavelet decomposition formula and decomposed into multiple scale and frequency components, each component corresponding to high-frequency and low-frequency components respectively;

[0070] Adaptive weighting adjustment: Use an adaptive frequency weighting function to adjust each component. Depending on the fault type, the function automatically increases or decreases the weight of certain frequency bands to highlight the signal features most useful for fault identification.

[0071] Comprehensive feature extraction: Finally, the feature extraction formula is used to summarize the signal energy at different scales and frequencies to form a comprehensive feature vector. The feature vector is input into the fault classifier or machine learning model for further analysis and assessment of the severity of the fault.

[0072] Further optimizing this technical solution, in step S3, the spatiotemporal coupling algorithm model utilizes the electromagnetic field data acquired by the optical CT measurement model and combines it with time series analysis to dynamically correlate the topological structure of the power system with the fault characteristics. The model captures the propagation path, speed, and impact range of the fault in the power system based on the spatiotemporal correlation coefficient and the dynamic adjacency matrix.

[0073] The time-space coupling algorithm model includes a time-space correlation calculation formula, a dynamic adjacency matrix formula, and a fault propagation path optimization formula.

[0074] To further optimize this technical solution, in the spatiotemporal coupling algorithm model:

[0075] The calculation formula for spatiotemporal correlation is as follows:

[0076] ;

[0077] in,

[0078] For in time and time difference Next, Hedi The spatiotemporal correlation coefficient between nodes;

[0079] For nodes In time The electromagnetic field strength;

[0080] For nodes The average value of the electromagnetic field during the time period;

[0081] is the time delay, which is used to capture the time effect of the fault signal propagating in the system;

[0082] The dynamic adjacency matrix formula is as follows:

[0083] ;

[0084] in,

[0085] is a dynamic adjacency matrix element used to represent the node and nodes Between time The connection status;

[0086] is the spatiotemporal correlation threshold, which is used to determine whether there is a fault correlation between two nodes;

[0087] For nodes and nodes The distance between the power grids;

[0088] The maximum allowed distance limits the spread of the fault's impact range;

[0089] The fault propagation path optimization formula is as follows:

[0090] ;

[0091] in,

[0092] is the optimal fault propagation path;

[0093] is the set of all possible fault propagation paths;

[0094] is the spatiotemporal correlation coefficient between nodes on the path;

[0095] The connection status between nodes ensures that the nodes on the path are related to each other.

[0096] To further optimize this technical solution, the spatiotemporal coupling algorithm model includes the following specific processes when used:

[0097] Real-time signal correlation analysis: After optical CT acquires electromagnetic field data, the correlation between nodes is calculated using a spatiotemporal correlation formula. Based on the signal propagation characteristics of the power system, the spatiotemporal relationship between nodes adjacent to the fault point is identified, forming a preliminary assessment of the fault impact area.

[0098] Dynamic network construction: Based on the calculated spatiotemporal correlation coefficients and system topology information, a dynamic adjacency matrix is ​​generated. Each element in the matrix represents the connection status between nodes at a specific moment. By setting thresholds and distance limits, irrelevant nodes are filtered out, forming a simplified network structure that focuses on fault propagation.

[0099] Fault propagation path tracing: Use the fault propagation path optimization formula to find the main propagation path of the fault in the system. The optimal path is used to analyze how the fault signal spreads within the system and help predict the possible impact of the fault on other critical equipment or areas.

[0100] To further optimize this technical solution, in step S7, the fault severity quantitative index system is constructed by including the following process:

[0101] Data collection and standardization: First, raw data on electromagnetic signal changes, power component health status, fault propagation speed, and duration are collected from the optical CT measurement model. Each indicator is normalized to eliminate dimensional differences between different dimensions.

[0102] Weight setting and adjustment: The weight of each indicator is set based on its importance in fault assessment. The weight is adjusted based on historical data and expert experience to ensure the model's adaptability to different types of faults. For example, if the propagation speed of a certain fault type has the greatest impact on system safety, the weight of the corresponding speed indicator will be appropriately increased.

[0103] Comprehensive score calculation: The standardized indicators are combined according to the preset weights and substituted into the fault severity score formula to calculate the final fault severity score. The higher the score, the more harmful the fault is to the power system and the higher the priority required for handling.

[0104] Compared with the existing technology, the present invention provides a method for evaluating the severity of power system faults based on optical CT measurement, which has the following beneficial effects:

[0105] This power system fault severity assessment method based on optical CT measurement significantly improves the accuracy of fault location and severity assessment by incorporating wavelet transforms, multiscale decomposition, a spatiotemporal coupling algorithm, and a comprehensive severity quantification indicator system. It overcomes the limitations of traditional technologies in fault signal capture and propagation path analysis, enabling real-time and comprehensive tracking of the dynamic changes of power system faults. Not only can it identify potential threats at the earliest stages of a fault, but it also provides a quantitative basis through comprehensive scoring, helping operations and maintenance personnel develop response strategies more quickly and effectively, thereby effectively improving the stability and reliability of the power grid. BRIEF DESCRIPTION OF THE DRAWINGS

[0106] Figure 1 This is a flow chart of a method for evaluating the severity of power system faults based on optical CT measurement proposed by the present invention;

[0107] Figure 2 This is a schematic diagram of the composition and flow of the optical CT measurement model in the power system fault severity assessment method based on optical CT measurement proposed by the present invention;

[0108] Figure 3 This is a schematic diagram of the composition and flow of a multi-scale decomposition model in a power system fault severity assessment method based on optical CT measurement proposed in the present invention;

[0109] Figure 4 This is a schematic diagram of the composition and flow of the time-space coupling algorithm model in the power system fault severity assessment method based on optical CT measurement proposed in the present invention;

[0110] Figure 5 This is a schematic diagram of the process of constructing a fault severity quantification index system in a power system fault severity assessment method based on optical CT measurement proposed by the present invention;

[0111] Figure 6 This is a diagram of a comprehensive scoring formula for fault severity in a power system fault severity assessment method based on optical CT measurement proposed by the present invention;

[0112] Figure 7 This is a schematic diagram of the weights of fault severity parameters in a power system fault severity assessment method based on optical CT measurement proposed by the present invention. DETAILED DESCRIPTION

[0113] The following will be combined with the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the embodiments described are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0114] Example:

[0115] See also Figure 1 A method for evaluating the severity of power system faults based on optical CT measurement is characterized by comprising the following steps:

[0116] S1. Constructing optical CT measurement model

[0117] By establishing an optical CT measurement model to obtain the electromagnetic field distribution in the power system, optical sensors are used to measure the spatial distribution characteristics of electromagnetic signals in real time without affecting the normal operation of the power grid. By installing optical CT sensor arrays at key nodes in the power system, multi-dimensional power signal data is captured. The signal data includes key information such as voltage waveform, current density, and electromagnetic wave propagation path.

[0118] In this embodiment, if Figure 2 As shown, in step S1, the constructed optical CT measurement model includes an electromagnetic field distribution equation and a signal reconstruction equation;

[0119] The electromagnetic field distribution equation is as follows:

[0120] ;

[0121] in,

[0122] For the spatial position and time The electromagnetic field distribution on

[0123] is the response function of the optical CT sensor, which depends on the wavelength and sensor azimuth ;

[0124] is the intensity function of the source electromagnetic signal, describing the current or voltage density in the power system over time. and spatial location distribution of

[0125] is the wave vector of the electromagnetic wave, describing the propagation direction;

[0126] is the distance vector between the sensor and the source signal;

[0127] is the angular frequency, which represents the frequency of the electromagnetic wave;

[0128] is the integration area, covering all measurement points in the power system.

[0129] Introduced the optical CT sensor response function This allows the model to adjust its measurement sensitivity based on the sensor's wavelength and azimuth. This allows the model to more accurately capture the characteristics of electromagnetic signals across different frequency bands, particularly during fault conditions, when electromagnetic signals in certain frequency bands of the power system can vary significantly. By adjusting the sensor's response function, abnormal signals can be more quickly identified in complex power systems.

[0130] Using Wave Vectors and the distance vector This improvement allows for tracking the propagation direction and velocity of fault signals, enabling more precise location of the fault source based on spatiotemporal coupling. This approach can help identify how electromagnetic fields propagate from a fault point in complex topologies and understand its impact on surrounding nodes.

[0131] The signal reconstruction equation is as follows:

[0132] ;

[0133] in,

[0134] is the reconstructed three-dimensional electromagnetic field intensity image;

[0135] is the weight coefficient, which represents the contribution of different optical CT sensors in the measurement process;

[0136] For the Sensors at location and time The measurement results on .

[0137] Introducing weight coefficient This coefficient is dynamically adjusted based on the performance of each optical CT sensor in different fault scenarios. The importance of sensor data can be adaptively adjusted based on the current power system status. For example, when a fault occurs in a certain area, the weight of the sensor closest to that area is automatically increased, thereby improving the accuracy of signal reconstruction and fault location.

[0138] Furthermore, when the optical CT measurement model is used, the following specific processes are included:

[0139] Real-time data acquisition: Utilize optical CT sensor arrays installed at key nodes of the power system to acquire electromagnetic field data in real time. The data is passed through the electromagnetic field distribution equation in the model. Processing is performed to capture the preliminary distribution of fault signals in the power system;

[0140] Fault signal reconstruction: Input the acquired data into the signal reconstruction equation ,Through the weight adjustment mechanism, the three-dimensional distribution image of the electromagnetic field at the time of the fault is reconstructed, and the fault point and the affected area are quickly identified;

[0141] Propagation path analysis: based on wave vector and the distance vector information, track the propagation path of fault signals, and predict their potential impact on other parts of the power system;

[0142] Adaptive response strategy generation: adjust the weight coefficients in the model based on real-time updated optical CT measurement data , generate the optimal fault response strategy, dynamically adjust the load distribution and the action of protection equipment in the power system, and reduce the impact of faults on the system.

[0143] S2. Multi-scale decomposition and feature extraction

[0144] After obtaining the original optical CT measurement signal, multi-scale decomposition is performed to extract the electromagnetic signal characteristics in different frequency bands. Wavelet transform technology is used to decompose the complex power signal into data at multiple scale levels, corresponding to high-frequency and low-frequency components, to capture the unique characteristics of different fault types in the time and frequency domains.

[0145] In this embodiment, in step S2, a multi-scale decomposition model is constructed based on wavelet transform technology, and time-frequency analysis technology is combined to achieve decomposition and feature extraction of signals in different frequency bands;

[0146] The model introduces nonlinear scale adjustment and frequency weighting mechanisms to improve the recognition accuracy and robustness of fault signal features;

[0147] like Figure 3 As shown, the multi-scale decomposition model includes a multi-scale wavelet decomposition formula, an adaptive frequency weighting function, and a feature extraction formula.

[0148] Furthermore, in the multi-scale decomposition model:

[0149] The multi-scale wavelet decomposition formula is as follows:

[0150] ;

[0151] in,

[0152] For scale and frequency The time-frequency coefficient under represents the energy distribution of the signal at different scales and frequency bands;

[0153] is the original optical CT measurement signal;

[0154] For scale and center frequency The mother wavelet function under the control of the fineness of signal decomposition;

[0155] Represents the conjugate complex number of the mother wavelet, which is used to capture the amplitude and phase information of the signal in the time-frequency domain.

[0156] The adaptive frequency weighting function is as follows:

[0157] ;

[0158] in,

[0159] is a frequency weighting function used to adjust the importance of signals in different frequency bands;

[0160] is the scale weight factor, which indicates the signal priority at different scales;

[0161] It is a frequency adjustment parameter used to control the attenuation speed of signals in different frequency bands;

[0162] is the target frequency, representing the fault signal frequency region of interest.

[0163] Scale weight factor Adaptive adjustments can be made based on the specific characteristics of optical CT measurement data. This means that for different types of faults (such as transient overcurrent and continuous harmonic distortion), the model will prioritize the appropriate scale level for analysis.

[0164] Through adaptive frequency weighting function Dynamically adjusting the signal energy in different frequency bands automatically selects important frequency regions based on the fault's characteristics, ignoring irrelevant or highly interfering signals. For example, when certain types of faults occur, the high-frequency components in the power system change more significantly. The model automatically weights these high-frequency signals, making the fault signature extraction more accurate.

[0165] The feature extraction formula is as follows:

[0166] ;

[0167] in,

[0168] is the final extracted feature vector, representing the comprehensive characteristics of the signal at different scales and frequencies;

[0169] and are discrete series of scale and frequency, respectively;

[0170] It represents the absolute energy value of the signal at a specific scale and frequency.

[0171] By weighting and aggregating the energy values ​​at different scales and frequencies, a high-dimensional feature vector is constructed. This feature vector is highly discriminative in fault analysis because it simultaneously considers the signal's changing characteristics in both the time and frequency domains. Traditional wavelet decomposition models are often limited to analyzing a single frequency band or time window, while this integrated analysis in the time and frequency domains allows for a comprehensive assessment of the evolution of fault signals.

[0172] When the multi-scale decomposition model is used, the following specific processes are included:

[0173] Signal decomposition and preprocessing: After obtaining the original optical CT measurement signal Then, it is input into the multi-scale wavelet decomposition formula and decomposed into multiple scale and frequency components, each component corresponding to high-frequency and low-frequency components respectively;

[0174] Adaptive weighting adjustment: Use an adaptive frequency weighting function to adjust each component. Depending on the fault type, the function automatically increases or decreases the weight of certain frequency bands to highlight the signal features most useful for fault identification.

[0175] Comprehensive feature extraction: Finally, the feature extraction formula is used to summarize the signal energy at different scales and frequencies to form a comprehensive feature vector. The feature vector is input into the fault classifier or machine learning model for further analysis and assessment of the severity of the fault.

[0176] S3. Time-space coupling fault location

[0177] Based on a spatiotemporal coupling algorithm model, the electromagnetic field data acquired by optical CT is combined with time series to accurately locate the fault point, track the development path of the fault within the power system in real time, and analyze its propagation speed and diffusion range. In this process, the model combines the topological information of the power system, dynamically associates the data of each node with neighboring nodes, and identifies the direction and range of possible fault spread.

[0178] In this embodiment, in step S3, the spatiotemporal coupling algorithm model uses the electromagnetic field data obtained by the optical CT measurement model and combines it with time series analysis to dynamically associate the topological structure of the power system with the fault characteristics. The model captures the propagation path, speed, and impact range of the fault in the power system based on the spatiotemporal correlation coefficient and the dynamic adjacency matrix.

[0179] like Figure 4 As shown, the time-space coupling algorithm model includes a time-space correlation calculation formula, a dynamic adjacency matrix formula, and a fault propagation path optimization formula.

[0180] Furthermore, in the space-time coupling algorithm model:

[0181] The calculation formula for spatiotemporal correlation is as follows:

[0182] ;

[0183] in,

[0184] For in time and time difference Next, Hedi The spatiotemporal correlation coefficient between nodes;

[0185] For nodes In time The electromagnetic field strength;

[0186] For nodes The average value of the electromagnetic field during the time period;

[0187] is the time delay, which is used to capture the time effect of the fault signal propagating in the system.

[0188] Calculate the correlation degree of fault signals between different nodes, taking into account not only the instantaneous value of the electromagnetic field signal, but also the time delay of signal propagation , thereby accurately capturing the dynamic propagation of fault signals in the network. This spatiotemporal correlation analysis can help identify how the fault signal gradually affects other nodes in the power system, thereby providing accurate fault location information.

[0189] The dynamic adjacency matrix formula is as follows:

[0190] ;

[0191] in,

[0192] is a dynamic adjacency matrix element used to represent the node and nodes Between time The connection status;

[0193] is the spatiotemporal correlation threshold, which is used to determine whether there is a fault correlation between two nodes;

[0194] For nodes and nodes The distance between the power grids;

[0195] The maximum allowed distance limits the spread of the fault impact range.

[0196] Dynamically adjust the connection status between nodes based on spatiotemporal correlation and network distance. By setting the spatiotemporal correlation threshold and distance restrictions ,The model can effectively filter out those irrelevant or distant nodes, ensuring that only those areas that are actually affected by the fault are analyzed.,This adaptive adjacency matrix design makes the model more accurate and efficient in fault analysis in complex power systems.

[0197] The fault propagation path optimization formula is as follows:

[0198] ;

[0199] in,

[0200] is the optimal fault propagation path;

[0201] is the set of all possible fault propagation paths;

[0202] is the spatiotemporal correlation coefficient between nodes on the path;

[0203] The connection status between nodes ensures that the nodes on the path are related to each other.

[0204] This optimization process seeks the optimal path for fault signal propagation within the power system. By maximizing the spatiotemporal correlation and connectivity between nodes along the path, it identifies the primary direction of fault propagation and the scope of impact. Compared to traditional static analysis methods, this path optimization mechanism can dynamically track the propagation of fault signals in real time, enabling operations and maintenance personnel to take quicker and more accurate countermeasures.

[0205] When the spatiotemporal coupling algorithm model is used, the following specific processes are included:

[0206] Real-time signal correlation analysis: After optical CT acquires electromagnetic field data, the correlation between nodes is calculated using a spatiotemporal correlation formula. Based on the signal propagation characteristics of the power system, the spatiotemporal relationship between nodes adjacent to the fault point is identified, forming a preliminary assessment of the fault impact area.

[0207] Dynamic network construction: Based on the calculated spatiotemporal correlation coefficients and system topology information, a dynamic adjacency matrix is ​​generated. Each element in the matrix represents the connection status between nodes at a specific moment. By setting thresholds and distance limits, irrelevant nodes are filtered out, forming a simplified network structure that focuses on fault propagation.

[0208] Fault propagation path tracking: using the fault propagation path optimization formula, find the main propagation path of the fault in the system, the optimal path is used to analyze how the fault signal spreads within the system, and helps to predict the impact of the fault on other critical devices or areas.

[0209] S4, dynamic network topology reconstruction

[0210] After the fault occurs, by updating the network topology model of the power system in real time, the relationship between the fault node and its surrounding nodes is reconstructed, based on the measurement results of the optical CT, the electromagnetic coupling relationship in the system is reconfigured, and the key path and element that may be affected by the fault are identified.

[0211] In this embodiment, this dynamic reconstruction process can quickly reflect the changes in system state, providing timely and accurate data support for severity assessment, so that the system can more efficiently analyze the potential impact on the entire network when facing multiple faults.

[0212] S5, fault impact propagation analysis

[0213] Impact propagation analysis is performed on the captured electromagnetic signals, by calculating the transmission path and speed of the fault in the network, the potential threat to other nodes and areas is evaluated, and the critical point of fault expansion and high-risk area are identified, providing reference for subsequent risk assessment and emergency response.

[0214] S6, power element health state prediction

[0215] Machine learning algorithms are used to predict the health status of key elements in the power system, combined with historical trend analysis of optical CT measurement data, to evaluate the failure probability of the element under different fault conditions.

[0216] In this embodiment, the stress level of the key element during the fault is predicted to determine whether it may fail in the short term, which is used to guide the development of maintenance plans and has a direct impact on the final assessment of fault severity.

[0217] S7, establishment of fault severity quantitative index

[0218] As shown in Figure 5-7 , a fault severity quantitative index system is established, based on the multi-dimensional parameters of electromagnetic signal change amplitude, power element health status, and fault impact propagation speed obtained by optical CT measurement, the parameters are standardized and given different weights, and a comprehensive severity score is calculated, the scoring system includes different types of fault scenarios, including instantaneous overcurrent, continuous overload, and line grounding, providing quantitative basis for risk level determination.

[0219] In this embodiment, the construction of the fault severity quantitative indicator system includes the following processes:

[0220] Data collection and standardization: First, raw data on electromagnetic signal changes, power component health status, fault propagation speed, and duration are collected from the optical CT measurement model. Each indicator is normalized to eliminate dimensional differences between different dimensions.

[0221] Weight setting and adjustment: The weight of each indicator is set based on its importance in fault assessment. The weight is adjusted based on historical data and expert experience to ensure the model's adaptability to different types of faults. For example, if the propagation speed of a certain fault type has the greatest impact on system safety, the weight of the corresponding speed indicator will be appropriately increased.

[0222] Comprehensive score calculation: The standardized indicators are combined according to the preset weights and substituted into the fault severity score formula to calculate the final fault severity score. The higher the score, the more harmful the fault is to the power system and the higher the priority required for handling.

[0223] The beneficial effects of the present invention are:

[0224] This power system fault severity assessment method based on optical CT measurement significantly improves the accuracy of fault location and severity assessment by incorporating wavelet transforms, multiscale decomposition, a spatiotemporal coupling algorithm, and a comprehensive severity quantification indicator system. It overcomes the limitations of traditional technologies in fault signal capture and propagation path analysis, enabling real-time and comprehensive tracking of the dynamic changes of power system faults. Not only can it identify potential threats at the earliest stages of a fault, but it also provides a quantitative basis through comprehensive scoring, helping operations and maintenance personnel develop response strategies more quickly and effectively, thereby effectively improving the stability and reliability of the power grid.

[0225] In the description of this specification, the reference terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" mean that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art can combine and combine different embodiments or examples described in this specification and features of different embodiments or examples without contradiction.

[0226] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A method for evaluating the severity of power system faults based on optical CT measurement, characterized in that: The following steps are involved: S1. Construct optical CT measurement model; By establishing an optical CT measurement model to obtain the electromagnetic field distribution in the power system, optical sensors are used to measure the spatial distribution characteristics of electromagnetic signals in real time without affecting the normal operation of the power grid. By installing optical CT sensor arrays at key nodes in the power system, multi-dimensional power signal data is captured. The signal data includes key information such as voltage waveform, current density, and electromagnetic wave propagation path. S2, multi-scale decomposition and feature extraction; After obtaining the original optical CT measurement signal, multi-scale decomposition is performed to extract the electromagnetic signal characteristics in different frequency bands. Wavelet transform technology is used to decompose the complex power signal into data at multiple scale levels, corresponding to high-frequency and low-frequency components, to capture the unique characteristics of different fault types in the time and frequency domains. S3, time-space coupled fault location; Based on a spatiotemporal coupling algorithm model, the electromagnetic field data acquired by optical CT is combined with time series to accurately locate the fault point, track the development path of the fault within the power system in real time, and analyze its propagation speed and diffusion range. In this process, the model combines the topological information of the power system to dynamically associate the data of each node with neighboring nodes, identifying the direction and range of possible fault spread. S4, dynamic network topology reconstruction; After a fault occurs, the network topology model of the power system is updated in real time to reconstruct the relationship between the fault node and its surrounding nodes. Based on the measurement results of optical CT, the electromagnetic coupling relationship within the system is reconfigured to identify the critical paths and components that may be affected by the fault. S5. Fault impact propagation analysis; Conduct impact propagation analysis on captured electromagnetic signals. By calculating the fault's transmission path and speed in the network, the potential threat to other nodes and regions is assessed, and critical points of fault expansion and high-risk areas are identified, providing a reference for subsequent risk assessment and emergency response. S6. Prediction of the health status of power components; Use machine learning algorithms to predict the health status of key components in power systems, combined with historical trend analysis of optical CT measurement data, to assess the failure probability of components under different fault conditions; S7. Establishment of quantitative indicators of fault severity; A quantitative indicator system for fault severity is established. Based on the multi-dimensional parameters of the electromagnetic signal change amplitude, power component health status, and fault impact propagation speed obtained by optical CT measurement, the parameters are standardized and assigned different weights to calculate a comprehensive severity score. The scoring system includes different types of fault scenarios, including instantaneous overcurrent, continuous overload, and line grounding, providing a quantitative basis for risk level determination.

2. The method for evaluating the severity of power system faults based on optical CT measurement according to claim 1, characterized in that: In step S1, the constructed optical CT measurement model includes an electromagnetic field distribution equation and a signal reconstruction equation; The electromagnetic field distribution equation is as follows: ； in, For the spatial position and time The electromagnetic field distribution on is the response function of the optical CT sensor, which depends on the wavelength and sensor azimuth ; is the intensity function of the source electromagnetic signal, describing the current or voltage density in the power system over time. and spatial location distribution of is the wave vector of the electromagnetic wave, describing the propagation direction; is the distance vector between the sensor and the source signal; is the angular frequency, which represents the frequency of the electromagnetic wave; is the integration area, covering all measurement points in the power system; The signal reconstruction equation is as follows: ； in, is the reconstructed three-dimensional electromagnetic field intensity image; is the weight coefficient, which represents the contribution of different optical CT sensors in the measurement process; For the Sensors at location and time The measurement results on .

3. The method for evaluating the severity of power system faults based on optical CT measurement according to claim 2, characterized in that: When the optical CT measurement model is used, the following specific processes are included: Real-time data acquisition: Utilize optical CT sensor arrays installed at key nodes of the power system to acquire electromagnetic field data in real time. The data is passed through the electromagnetic field distribution equation in the model. Processing is performed to capture the preliminary distribution of fault signals in the power system; Fault signal reconstruction: Input the acquired data into the signal reconstruction equation ,Through the weight adjustment mechanism, the three-dimensional distribution image of the electromagnetic field at the time of the fault is reconstructed, and the fault point and the affected area are quickly identified; Propagation path analysis: based on wave vector and the distance vector information, track the propagation path of fault signals, and predict their potential impact on other parts of the power system; Adaptive response strategy generation: adjust the weight coefficients in the model based on real-time updated optical CT measurement data , generate the optimal fault response strategy, dynamically adjust the load distribution and the action of protection equipment in the power system, and reduce the impact of faults on the system.

4. The method for evaluating the severity of power system faults based on optical CT measurement according to claim 1, characterized in that: In step S2, a multi-scale decomposition model is constructed based on wavelet transform technology, and time-frequency analysis technology is combined to achieve decomposition and feature extraction of signals in different frequency bands; The model introduces nonlinear scale adjustment and frequency weighting mechanisms to improve the recognition accuracy and robustness of fault signal features; The multi-scale decomposition model includes a multi-scale wavelet decomposition formula, an adaptive frequency weighting function, and a feature extraction formula.

5. The method for evaluating the severity of power system faults based on optical CT measurement according to claim 4, characterized in that: In the multi-scale decomposition model: The multi-scale wavelet decomposition formula is as follows: ； in, For scale and frequency The time-frequency coefficient under represents the energy distribution of the signal at different scales and frequency bands; is the original optical CT measurement signal; For scale and center frequency The mother wavelet function under the control of the fineness of signal decomposition; Represents the conjugate complex number of the mother wavelet, which is used to capture the amplitude and phase information of the signal in the time-frequency domain; The adaptive frequency weighting function is as follows: ； in, is a frequency weighting function used to adjust the importance of signals in different frequency bands; is the scale weight factor, which indicates the signal priority at different scales; It is a frequency adjustment parameter used to control the attenuation speed of signals in different frequency bands; is the target frequency, representing the fault signal frequency region of interest; The feature extraction formula is as follows: ； in, is the final extracted feature vector, representing the comprehensive characteristics of the signal at different scales and frequencies; and are discrete series of scale and frequency, respectively; It represents the absolute energy value of the signal at a specific scale and frequency.

6. The method for evaluating the severity of power system faults based on optical CT measurement according to claim 4, characterized in that: When the multi-scale decomposition model is used, the following specific processes are included: Signal decomposition and preprocessing: After obtaining the original optical CT measurement signal Then, it is input into the multi-scale wavelet decomposition formula and decomposed into multiple scale and frequency components, each component corresponding to high-frequency and low-frequency components respectively; Adaptive weighting adjustment: Use an adaptive frequency weighting function to adjust each component. Depending on the fault type, the function automatically increases or decreases the weight of certain frequency bands to highlight the signal features most useful for fault identification. Comprehensive feature extraction: Finally, the feature extraction formula is used to summarize the signal energy at different scales and frequencies to form a comprehensive feature vector. The feature vector is input into the fault classifier or machine learning model for further analysis and assessment of the severity of the fault.

7. The method for evaluating the severity of power system faults based on optical CT measurement according to claim 1, characterized in that: In step S3, the spatiotemporal coupling algorithm model uses the electromagnetic field data obtained by the optical CT measurement model and combines it with time series analysis to dynamically associate the topology of the power system with the fault characteristics. The model captures the propagation path, speed, and impact range of the fault in the power system based on the spatiotemporal correlation coefficient and the dynamic adjacency matrix. The time-space coupling algorithm model includes a time-space correlation calculation formula, a dynamic adjacency matrix formula, and a fault propagation path optimization formula.

8. The method for evaluating the severity of power system faults based on optical CT measurement according to claim 7, characterized in that: In the space-time coupling algorithm model: The calculation formula for spatiotemporal correlation is as follows: ； in, For in time and time difference Next, Hedi The spatiotemporal correlation coefficient between nodes; For nodes In time The electromagnetic field strength; For nodes The average value of the electromagnetic field during the time period; is the time delay, which is used to capture the time effect of the fault signal propagating in the system; The dynamic adjacency matrix formula is as follows: ； in, is a dynamic adjacency matrix element used to represent the node and nodes Between time The connection status; is the spatiotemporal correlation threshold, which is used to determine whether there is a fault correlation between two nodes; For nodes and nodes The distance between the power grids; The maximum allowed distance limits the spread of the fault's impact range; The fault propagation path optimization formula is as follows: ； in, is the optimal fault propagation path; is the set of all possible fault propagation paths; is the spatiotemporal correlation coefficient between nodes on the path; The connection status between nodes ensures that the nodes on the path are related to each other.

9. The method for evaluating the severity of power system faults based on optical CT measurement according to claim 8, characterized in that: When the spatiotemporal coupling algorithm model is used, the following specific processes are included: Real-time signal correlation analysis: After optical CT acquires electromagnetic field data, the correlation between nodes is calculated using a spatiotemporal correlation formula. Based on the signal propagation characteristics of the power system, the spatiotemporal relationship between nodes adjacent to the fault point is identified, forming a preliminary assessment of the fault impact area. Dynamic network construction: Based on the calculated spatiotemporal correlation coefficients and system topology information, a dynamic adjacency matrix is ​​generated. Each element in the matrix represents the connection status between nodes at a specific moment. By setting thresholds and distance limits, irrelevant nodes are filtered out, forming a simplified network structure that focuses on fault propagation. Fault propagation path tracing: Use the fault propagation path optimization formula to find the main propagation path of the fault in the system. The optimal path is used to analyze how the fault signal spreads within the system and help predict the possible impact of the fault on other critical equipment or areas.

10. The method for evaluating the severity of power system faults based on optical CT measurement according to claim 1, characterized in that: In step S7, the fault severity quantitative index system is constructed by including the following process: Data collection and standardization: First, raw data on electromagnetic signal changes, power component health status, fault propagation speed, and duration are collected from the optical CT measurement model. Each indicator is normalized to eliminate dimensional differences between different dimensions. Weight setting and adjustment: The weight of each indicator is set based on its importance in fault assessment. The weight is adjusted based on historical data and expert experience to ensure the model's adaptability to different types of faults. For example, if the propagation speed of a certain fault type has the greatest impact on system safety, the weight of the corresponding speed indicator will be appropriately increased. Comprehensive score calculation: The standardized indicators are combined according to the preset weights and substituted into the fault severity score formula to calculate the final fault severity score. The higher the score, the more harmful the fault is to the power system and the higher the priority required for handling.

Images