Traveling wave fault location-based power grid fault location system and method

Abstract

The invention relates to the technical field of fault analysis, in particular to a power grid fault positioning system and method based on traveling wave ranging, and the system comprises a signal capturing and analyzing module, a dynamic tracking and recognizing module, a nonlinear feature analyzing module and a fault positioning and reconstructing module. According to the invention, each instantaneous point of the signal is tracked carefully, so that the positioning of the fault is more accurate, the actual influence of the fault on the stability of the power grid can be evaluated more comprehensively by extracting the nonlinear characteristic value in the traveling wave signal and analyzing the dynamic characteristics of the fault signal through data, and the improvement of the propagation path of the signal is tracked reversely. According to the method, the positioning of a fault point not only depends on theoretical data, but also is checked by combining actual geographic data, so that the accuracy of fault source positioning is greatly improved, a method integrating signal analysis and geographic data provides a more systematic and comprehensive fault response mechanism, the maintenance operation of a power system is optimized, and the reliability of the power system is improved. The power supply interruption time caused by a fault is remarkably shortened.

Description

Technical Field

[0001] The present invention relates to the technical field of fault analysis, and in particular to a power grid fault location system and method based on traveling wave ranging. Background Art

[0002] The field of fault analysis technology involves a series of methods and equipment for detecting, analyzing and locating faults in power systems. It includes fault detection in power transmission and distribution networks, and accurate diagnosis of fault types, locations and possible causes. Technical means usually include signal processing, data analysis and simulation technology, as well as real-time monitoring systems, which are of great significance to improving the reliability, safety and efficiency of the power grid, and can quickly respond to and handle problems in the power grid, reducing power outage time and maintenance costs.

[0003] Among them, the power grid fault location system is mainly used to identify and locate fault points in the power grid, by monitoring the current and voltage changes of the power grid and analyzing the data to determine the specific location of the fault. The uses include accelerating fault response time, optimizing maintenance operations, and reducing the power supply interruption time caused by faults, thereby improving the overall stability and efficiency of the power system. It is particularly important for modern power infrastructure and supports faster fault recovery and enhanced network management.

[0004] Current technologies mainly rely on traditional signal processing and data analysis methods, which often lack sufficient sensitivity and accuracy when processing complex or nonlinear fault signals, resulting in delays in fault diagnosis and reduced accuracy. In addition, existing technologies often rely only on simulation technology and theoretical models in fault point location, without fully utilizing actual field data, which may lead to misjudgment in complex power grid environments. For example, inaccurate fault location may cause unnecessary maintenance work, increase operation and maintenance costs, and take more time to recover from the fault, affecting the overall efficiency and reliability of the power grid. Summary of the invention

[0005] The purpose of the present invention is to solve the shortcomings existing in the prior art and to propose a power grid fault location system and method based on traveling wave ranging.

[0006] In order to achieve the above object, the present invention adopts the following technical solution: A power grid fault location system based on traveling wave ranging includes:

[0007] The signal capture and analysis module continuously collects traveling wave signals through ADC, measures the signal peak height, time interval and waveform change, timestamps and preliminarily classifies the data, identifies key events in the signal, and generates key morphological features;

[0008] The dynamic tracking and identification module uses the key morphological features to track the signal waveform in detail, detect each instantaneous point of the traveling wave signal, analyze the correspondence between the signal change point and the time series, identify the traveling wave change pattern based on the corresponding relationship analysis results, evaluate the abnormal behavior of the traveling wave signal, analyze the change trend of the signal waveform, and generate dynamic offset prediction analysis results;

[0009] The nonlinear characteristic analysis module extracts the nonlinear characteristic value in the traveling wave signal according to the dynamic offset prediction analysis result, calculates the nonlinear dynamic index of the traveling wave signal, analyzes the nonlinear dynamic characteristics of the fault signal according to the calculation result, evaluates the impact on the stability of the power grid, and outputs an enhanced offset signal;

[0010] The fault location and reconstruction module reversely tracks the propagation path and time delay of the traveling wave signal based on the enhanced offset signal, analyzes the possible location of the fault point, determines the accurate location of the fault source by checking with the actual geographic data, and generates a power grid fault location result.

[0011] As a further solution of the present invention, the step of acquiring the key morphological features is:

[0012] The traveling wave signal is continuously monitored by the ADC device, and the voltage value and corresponding acquisition time of each data point are continuously recorded to generate a raw waveform data set with a time stamp;

[0013] Based on the original waveform data set with time stamps, each peak is identified, and the corresponding timestamp is recorded, and the signal is preliminarily classified to obtain a classified peak feature data set;

[0014] Based on the classified peak feature data set, the formula is adopted:

[0015]

[0016] Calculate the total peak height H and generate key morphological features, where p i represents the height of the ith peak, t i It represents the time interval between the ith peak and the previous peak, and n represents the total number of peaks.

[0017] As a further solution of the present invention, the analysis steps of the corresponding relationship are:

[0018] Based on the key morphological features, segmentation processing is performed in time sequence, and the segmented waveform data is screened by numerical range to generate a signal waveform data set with time tags;

[0019] The change point detection is performed on the signal waveform data set with time tags using the formula:

[0020]

[0021] Calculate the coefficient of variation C of consecutive signal points i , identify each instantaneous change point and obtain the instantaneous signal change point set, where S i and S i+1 is the value of two consecutive signal points;

[0022] The instantaneous signal change point set is used to arrange the change points in sequence according to the time labels, mark the time coordinate of each change point, and generate a corresponding relationship result between the change point and the time series.

[0023] As a further solution of the present invention, the steps of obtaining the dynamic offset prediction analysis result are:

[0024] Based on the corresponding relationship between the change points and the time series, the change points in the time series are analyzed point by point, and the time interval Δt between consecutive change points is calculated. i and amplitude difference ΔA i , and extract the fluctuation amplitude and fluctuation direction of each change point, summarize the change characteristics of continuous points, and generate a traveling wave change feature set;

[0025] The traveling wave variation feature set is called, and the fluctuation feature is numerically evaluated by a normalization method, using the formula:

[0026]

[0027] Calculate the abnormal deviation value D of the overall fluctuation and generate the abnormal behavior result, where ΔA i represents the fluctuation amplitude of the ith point, is the average value of all fluctuation amplitudes, and N represents the total number of fluctuation points;

[0028] Based on the abnormal behavior results and the traveling wave change feature set, the time period and amplitude change characteristics corresponding to the abnormal area are extracted segment by segment, the increase and decrease trends are analyzed, and the offset direction of the overall waveform trend is determined to generate dynamic offset prediction analysis results.

[0029] As a further solution of the present invention, the calculation steps of the nonlinear dynamic index are:

[0030] Using the dynamic shift prediction analysis results, deep spectrum analysis is performed to extract signal components with significant nonlinear characteristics and generate a traveling wave signal nonlinear characteristic value sequence;

[0031] Based on the nonlinear characteristic value sequence of the traveling wave signal, statistical analysis is performed to determine the mean, standard deviation and kurtosis coefficient, characterize the dynamic change trend of the signal, and generate statistical dynamic parameters of the traveling wave signal;

[0032] Combining the nonlinear characteristic value sequence of the traveling wave signal and the statistical dynamic parameters of the traveling wave signal, the formula is adopted:

[0033]

[0034] The nonlinear dynamic index NDI of the traveling wave signal is calculated, where f i represents the i-th nonlinear eigenvalue in the traveling wave signal, p i Represents the corresponding statistical dynamic parameter, μ represents the mean of the nonlinear eigenvalue sequence, and m represents the total number of nonlinear eigenvalues.

[0035] As a further solution of the present invention, the step of acquiring the enhanced offset signal is:

[0036] Based on the nonlinear dynamic index of the traveling wave signal, the nonlinear dynamic characteristics of the fault signal are analyzed, the potential fault mode matching the current index is identified, the fault area and type are located, and the nonlinear dynamic characteristics analysis result of the fault signal is generated;

[0037] Using the nonlinear dynamic characteristic analysis results of the fault signal, simulating the power grid response according to the severity of the fault characteristics, evaluating the potential impact of the fault on the power grid stability, and generating a power grid stability impact assessment result;

[0038] According to the grid stability impact assessment result, the signal processing parameters are adjusted to optimize the grid's responsiveness and fault tolerance to potential faults, and an enhanced offset signal is output.

[0039] As a further solution of the present invention, the steps of obtaining the power grid fault location result are:

[0040] Using the enhanced offset signal, reverse tracking is performed to determine the propagation path and time delay of the traveling wave signal, and by analyzing the speed and waveform characteristics of the traveling wave signal, signal propagation path data is obtained;

[0041] Based on the signal propagation path data, combined with the power grid topology and equipment location information, data fusion and analysis are performed to identify potential fault point locations and generate a list of candidate fault point locations;

[0042] The candidate fault location list is used to check with actual geographic and equipment operation data, and the exact location of the fault source is determined through data comparison and analysis to generate a power grid fault location result.

[0043] A method for locating power grid faults based on traveling wave ranging comprises the following steps:

[0044] S1: Continuously monitor the traveling wave signal through the ADC device, continuously record the voltage value and corresponding acquisition time of each data point, identify each peak, and record the corresponding timestamp, calculate the total height value of the peak, and generate key morphological features;

[0045] S2: Based on the key morphological features, segment processing is performed in time sequence, change point detection is performed, the change coefficients of continuous signal points are calculated, each instantaneous change point is identified, the time coordinates of each change point are marked, and the corresponding relationship between the change point and the time series is generated;

[0046] S3: Based on the corresponding relationship between the change point and the time series, the change points in the time series are analyzed point by point, the fluctuation characteristics are numerically evaluated by the normalization method, the time period and amplitude change characteristics corresponding to the abnormal area are extracted and the increase and decrease trends are analyzed, the deviation direction of the overall waveform trend is determined, and the dynamic deviation prediction analysis results are generated;

[0047] S4: using the dynamic offset prediction analysis results, extracting the signal components with significant nonlinear characteristics, performing statistical analysis to characterize the dynamic change trend of the signal, and calculating and outputting the nonlinear dynamic index of the traveling wave signal;

[0048] S5: Based on the nonlinear dynamic index of the traveling wave signal, analyze the nonlinear dynamic characteristics of the fault signal, locate the fault area and type, evaluate the potential impact of the fault on the stability of the power grid, adjust the signal processing parameters, and output an enhanced offset signal;

[0049] S6: Use the enhanced offset signal to determine the propagation path and time delay of the traveling wave signal, analyze the speed and waveform characteristics of the traveling wave signal, combine it with the power grid topology and equipment location information, identify the location of the potential fault point, check it with the actual geographic and equipment operation data, and generate a power grid fault location result.

[0050] Compared with the prior art, the advantages and positive effects of the present invention are:

[0051] In the present invention, by carefully tracking each instantaneous point of the signal, the fault location can be more accurately located. By extracting the nonlinear eigenvalues ​​in the traveling wave signal and using data to analyze the dynamic characteristics of the fault signal, the actual impact of the fault on the stability of the power grid can be more comprehensively evaluated. The improvement of the reverse tracking signal propagation path makes the location of the fault point not only rely on theoretical data, but also combined with actual geographic data for verification, thereby greatly improving the accuracy of fault source location. The method of integrating signal analysis and geographic data provides a more systematic and comprehensive fault response mechanism, optimizes the maintenance operation of the power system, and significantly shortens the power supply interruption time caused by the fault. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1is a system flow chart of the present invention;

[0053] Figure 2 It is a flow chart for obtaining the key morphological features of the present invention;

[0054] Figure 3 It is an analysis flow chart of the corresponding relationship of the present invention;

[0055] Figure 4 A flowchart for obtaining dynamic offset prediction and analysis results of the present invention;

[0056] Figure 5 It is a calculation flow chart of the nonlinear dynamic index of the present invention;

[0057] Figure 6 The present invention is a flowchart for obtaining an enhanced offset signal;

[0058] Figure 7 This is a flow chart of obtaining the power grid fault location result of the present invention. DETAILED DESCRIPTION

[0059] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0060] In the description of the present invention, it should be understood that the terms "length", "width", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside" and the like indicate positions or positional relationships based on the positions or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, in the description of the present invention, "multiple" means two or more, unless otherwise clearly and specifically defined.

[0061] See also Figure 1 , a power grid fault location system based on traveling wave ranging includes:

[0062] The signal capture and analysis module continuously collects traveling wave signals through ADC, measures the signal peak height, time interval and waveform change, timestamps and preliminarily classifies the data, identifies key events in the signal, and generates key morphological features;

[0063] The dynamic tracking and identification module uses key morphological features to track the signal waveform in detail, detect each instantaneous point of the traveling wave signal, analyze the correspondence between the signal change point and the time series, identify the traveling wave change pattern based on the corresponding relationship analysis results, evaluate the abnormal behavior of the traveling wave signal, analyze the change trend of the signal waveform, and generate dynamic offset prediction analysis results;

[0064] The nonlinear characteristic analysis module extracts the nonlinear characteristic value in the traveling wave signal according to the dynamic offset prediction analysis results, calculates the nonlinear dynamic index of the traveling wave signal, analyzes the nonlinear dynamic characteristics of the fault signal according to the calculation results, evaluates the impact on the stability of the power grid, and outputs an enhanced offset signal;

[0065] The fault location and reconstruction module reversely tracks the propagation path and time delay of the traveling wave signal based on the enhanced offset signal, analyzes the possible location of the fault point, determines the accurate location of the fault source by checking with the actual geographic data, and generates the power grid fault location result.

[0066] Key morphological features include peak height parameters, time interval indicators and waveform change data; dynamic offset prediction analysis results include pattern recognition data, abnormal behavior indicators and trend change analysis records; enhanced offset signals include nonlinear feature extraction records, dynamic response indicators and stability impact evaluation results; power grid fault location results include fault location information, time delay calculation results and geographic matching verification results.

[0067] See also Figure 2 , the steps to obtain the key morphological features are:

[0068] The traveling wave signal is continuously monitored by the ADC device, and the voltage value and corresponding acquisition time of each data point are continuously recorded to generate a raw waveform data set with a time stamp;

[0069] The traveling wave signal is continuously monitored by a high-precision ADC device, and the voltage value of each data point and the corresponding precise acquisition time are continuously recorded in each acquisition cycle. The process relies on a synchronized clock system to ensure the accuracy of the timestamp. The instantaneous changes of the signal are captured by high-speed sampling, thereby ensuring the integrity and accuracy of the waveform data. The raw data constitutes a detailed signal waveform diagram, providing basic data for subsequent analysis. In addition, the data point is recorded using a specific data recording format to ensure the readability of the data and the convenience of subsequent processing. The generated time-stamped raw waveform data set contains not only the voltage value of each data point, but also the corresponding time stamp, providing key time and voltage information for subsequent data analysis.

[0070] Based on the original waveform data set with time tags, each peak is identified and the corresponding timestamp is recorded, and the signal is preliminarily classified to obtain a classified peak feature data set;

[0071] In the process of identifying peaks and their timestamps from the original waveform data set, signal processing techniques are used to analyze and process each waveform. First, a voltage threshold is set to identify all signal points that exceed this threshold, and then a sliding window method is applied to determine the exact position and height of each peak. In addition, by comparing the time interval between two adjacent peaks and counting the frequency of peaks, the characteristics of the peaks are used to classify the signal, such as dividing the waveform into high-frequency and low-frequency categories. The classification process is based on the physical properties of the peaks, such as height and time interval. In this way, not only are the peaks identified, but also the peaks are preliminarily classified, providing a preprocessed classification data set for further data processing and analysis, reflecting the basic behavior and characteristics of the signal.

[0072] Based on the classified peak feature data set, the formula is used:

[0073]

[0074] Calculate the total peak height H and generate key morphological features, where p i represents the height of the ith peak, t i represents the time interval between the ith peak and the previous peak, and n represents the total number of peaks;

[0075] In one acquisition cycle, three peaks are monitored, with heights of p1 = 5 volts, p2 = 10 volts, and p3 = 15 volts, respectively. The corresponding time intervals are t1 = 1 second, t2 = 4 seconds, and t3 = 9 seconds. Calculation is performed according to the formula:

[0076]

[0077] The results show that, taking into account the effect of time interval on the peak, the weighted total peak height is 15, which can be used to evaluate the overall activity or energy intensity of the peak in the signal, and can help quantify the key events and waveform changes in the signal, thus providing a quantitative basis for further signal processing and analysis.

[0078] See also Figure 3 , the analysis steps of the corresponding relationship are:

[0079] Based on key morphological features, segmentation processing is performed in chronological order, and the segmented waveform data is screened for numerical range to generate a signal waveform data set with time tags;

[0080] Based on the key morphological features, the input original signal waveform data is first segmented in chronological order, and the key morphological features of the original waveform such as amplitude, frequency, period and zero crossing are extracted in each period of time. The segmented waveform data is screened in the numerical range to eliminate abnormal data and noise interference signals. The specific implementation is to set upper and lower limit thresholds for the amplitude and period of each segment of waveform data, and eliminate data points that do not meet the threshold range. Then, the key features of the data in the effective band are secondary extracted, including local extreme points, waveform slope changes, and the time difference between the peak and the trough. Through the extraction of features, the data is synchronously paired with the time label, and all the extracted morphological features are integrated and stored in chronological order to form a signal waveform data set with time labels.

[0081] For the signal waveform data set with time tags, change point detection is performed using the formula:

[0082]

[0083] Calculate the coefficient of variation C of consecutive signal points i , identify each instantaneous change point and obtain the instantaneous signal change point set, where S i and S i+1 is the value of two consecutive signal points;

[0084] Given the known signal sampling data, the collected continuous signal point values ​​are as follows:

[0085] S1=5, S2=8, S3=12, S4=11

[0086] Calculate the first coefficient of variation C1:

[0087]

[0088] Calculate the second coefficient of variation C2:

[0089]

[0090] Calculate the third coefficient of variation C3:

[0091]

[0092] C1=0.2307: It means that between the first change point (from S1 to S2), the signal has a significant upward trend and the degree of change is high.

[0093] C2=0.2: indicates that between the second change point (from S2 to S3), the signal continues to rise, but the change amplitude is slightly smaller than before.

[0094] C3=-0.0435: indicates that between the third change point (from S3 to S4), the signal drops slightly, the degree of change is small, and the direction reverses to negative.

[0095] Using the instantaneous signal change point set, the change points are arranged in sequence according to the time labels, the time coordinates of each change point are marked, and the corresponding relationship between the change point and the time series is generated;

[0096] Using the signal change point set obtained above, the change points are arranged in sequence according to the time labels and the specific time coordinates of each change point are marked. Then, the time points corresponding to the change points are read one by one, the time intervals between adjacent change points are calculated, and the time difference between every two change points is determined by point-by-point comparison. The change point areas with frequent fluctuations in continuous time intervals are screened out. On this basis, each change point and its time point are paired, and a preliminary analysis of the overall change trend of the waveform is performed. Specifically, all key change points are marked on the time axis and the amplitudes of these change points are connected. The increase and decrease slope and change amplitude of the overall trend are calculated, and detailed results including the relationship between the change points and the time series are generated.

[0097] See also Figure 4 , the steps to obtain the dynamic offset prediction analysis results are:

[0098] Based on the corresponding relationship between the change points and the time series, the change points in the time series are analyzed point by point, and the time interval Δt between consecutive change points is calculated. i and amplitude difference ΔA i , and extract the fluctuation amplitude and fluctuation direction of each change point, summarize the change characteristics of continuous points, and generate a traveling wave change feature set;

[0099] Based on the corresponding relationship between the change points and the time series, the time interval and amplitude difference between consecutive change points are calculated by analyzing the change points in the time series point by point. First, the time mark of each change point is extracted, and the time difference Δt between adjacent change points is calculated. i and the amplitude difference ΔA i (Δt i =t i+1 -t i , ΔA i =|A i+1 -A i |, where t i and t i+1 Respectively represent the time points of the i-th and i+1-th change points, A i and A i+1Represents the corresponding signal amplitude), and then compares the change amplitudes between these change points one by one, and calculates the fluctuation amplitude and fluctuation direction of each change point. By calculating the change trend of the fluctuation amplitude, the fluctuation characteristics of each period of time are obtained, and finally the key features representing the traveling wave mode are extracted. The traveling wave change feature set formed after analysis contains information such as time interval, amplitude difference and fluctuation amplitude, providing data support for subsequent abnormal behavior analysis and dynamic prediction.

[0100] The traveling wave variation feature set is called, and the fluctuation characteristics are numerically evaluated by the normalization method, using the formula:

[0101]

[0102] Calculate the abnormal deviation value D of the overall fluctuation and generate the abnormal behavior result, where ΔA i represents the fluctuation amplitude of the ith point, is the average value of all fluctuation amplitudes, and N represents the total number of fluctuation points;

[0103] The extracted fluctuation range difference data is: ΔA1=5, ΔA2=8, ΔA3=6, ΔA4=7, then calculate is the average of these values:

[0104]

[0105] Substitute the formula to calculate D:

[0106]

[0107] The calculated deviation value D=1 indicates the degree of deviation between these fluctuation amplitudes and the average fluctuation amplitude, which can be used to determine the abnormality of the traveling wave signal. A larger D value indicates that the signal fluctuation is relatively abnormal, and there may be large changes or abnormal behavior; a smaller D value indicates that the fluctuation is relatively stable.

[0108] Based on the abnormal behavior results and the traveling wave change feature set, the time period and amplitude change characteristics corresponding to the abnormal area are extracted segment by segment, the increase and decrease trends are analyzed, and the deviation direction of the overall waveform trend is determined to generate dynamic deviation prediction analysis results;

[0109] The abnormal behavior results and the wave change feature set are called to analyze the time series data in the abnormal behavior area. First, for each change point in the abnormal area, the fluctuation amplitude and time series change trend of the change point are calculated to further determine the distribution of each change point in time, and compare it with the fluctuation characteristics of the normal area. Through comparison, the time period with large fluctuations can be identified and the offset direction of the waveform can be determined. Next, data integration is performed based on the features, the fluctuation amplitude of the abnormal area is compared with the fluctuation characteristics of the normal area, and the changes in the overall trend are analyzed. Finally, the dynamic offset prediction analysis results are generated based on the analysis results, providing a quantitative basis for subsequent signal trend prediction and anomaly detection.

[0110] See also Figure 5 , the calculation steps of nonlinear dynamic index are:

[0111] Using the dynamic offset prediction analysis results, deep spectrum analysis is performed to extract signal components with significant nonlinear characteristics and generate a traveling wave signal nonlinear characteristic value sequence.

[0112] In order to accurately extract signal components with significant nonlinear characteristics from traveling wave signals, signal acquisition is first performed to convert the real-time monitored power grid traveling wave signals into digital format, and the fast Fourier transform (FFT) is used to perform spectrum analysis on these signals. This process involves converting the time domain signal into the frequency domain in order to better analyze the signal frequency components; the signal is then processed using a designed high-pass filter, which can effectively remove noise components with frequencies below the set threshold and retain the high-frequency parts that are more valuable for analysis; then, the signal is restored from the frequency domain to the time domain through an inverse Fourier transform, and the signal format for time series analysis can be regained; finally, from the restored signal, the nonlinear eigenvalues ​​are accurately extracted through the detection of the local maximum value of the amplitude, and these values ​​are combined into a sequence for the next step of dynamic analysis and processing to generate a nonlinear eigenvalue sequence of the traveling wave signal.

[0113] Based on the nonlinear characteristic value sequence of the traveling wave signal, statistical analysis is performed to determine the mean, standard deviation and kurtosis coefficient, characterize the dynamic change trend of the signal, and generate the statistical dynamic parameters of the traveling wave signal;

[0114] After extracting the nonlinear eigenvalue sequence of the traveling wave signal, in order to have a deeper understanding of the dynamic characteristics of the signal, a detailed statistical analysis of the eigenvalues ​​is performed, including the calculation of the mean, standard deviation and kurtosis coefficient of the sequence; first, the mean is calculated by summing up all the eigenvalues ​​and dividing by the number of eigenvalues. This value represents the average level of the eigenvalues ​​of the entire signal; then, in order to measure the degree of fluctuation of the eigenvalues, the standard deviation is calculated by first finding the square of the difference between each eigenvalue and the mean, then summing these squared values, and finally taking the square root of the sum; the kurtosis coefficient is calculated to evaluate the sharpness of the eigenvalue distribution. It is obtained by summing the fourth power of the difference between all eigenvalues ​​and the mean and dividing it by the fourth power of the standard deviation, thus generating the statistical dynamic parameters of the traveling wave signal.

[0115] The nonlinear eigenvalue sequence of the traveling wave signal and the statistical dynamic parameters of the traveling wave signal are integrated, and the formula is adopted:

[0116]

[0117] The nonlinear dynamic index NDI of the traveling wave signal is calculated, where f i represents the i-th nonlinear eigenvalue in the traveling wave signal, p i represents the corresponding statistical dynamic parameter, μ represents the mean of the nonlinear eigenvalue sequence, and m represents the total number of nonlinear eigenvalues;

[0118] There are three eigenvalues ​​f1=2.0, f2=2.5, f3=3.0, the corresponding statistical parameters p1=1.1, p2=1.2, p3=1.3, and the mean μ=2.5. Substitute them into the formula for calculation:

[0119]

[0120] but:

[0121]

[0122] The results show that the nonlinear dynamic index is 3.22, indicating that the nonlinear dynamic characteristics are quite significant, indicating that the traveling wave signal has significant nonlinear behavior.

[0123] See also Figure 6 , the steps to obtain the enhanced offset signal are:

[0124] Based on the nonlinear dynamic index of the traveling wave signal, the nonlinear dynamic characteristics of the fault signal are analyzed, the potential fault mode matching the current index is identified, the fault area and type are located, and the nonlinear dynamic characteristics analysis results of the fault signal are generated;

[0125] By analyzing the nonlinear dynamic indicators of the traveling wave signal in detail, fault pattern matching and fault location are performed. First, the traveling wave signal data is imported into the fault analysis system. The system uses a pre-trained decision tree classifier to analyze the nonlinear dynamic indicators and identify patterns that match the known fault patterns. The classifier classifies the fault indicators according to their size, frequency and their changing trends. Subsequently, the spatial attributes and timestamp information of the signal are used to map it to the power grid model through the geographic information system. The shortest path algorithm is used to calculate the propagation path of the signal from the location of occurrence to the detection point, thereby accurately determining the location and type of the fault. This process ensures the accuracy and response speed of fault location, and ultimately generates detailed analysis results of the nonlinear dynamic characteristics of the fault signal.

[0126] Using the nonlinear dynamic characteristics analysis results of the fault signal, simulate the grid response according to the severity of the fault characteristics, evaluate the potential impact of the fault on the grid stability, and generate the grid stability impact assessment results;

[0127] After obtaining the nonlinear dynamic characteristics of the fault signal, the impact on the stability of the power grid is evaluated. This evaluation uses specific power grid simulation software to simulate the current and voltage response of the power grid when a fault occurs, based on the input fault type and location parameters, and calculate the voltage and current changes at each node of the power grid, thereby predicting the impact range of the fault. Based on the simulation results, the chain reactions that may be caused by the fault are analyzed, such as voltage sag and frequency fluctuation. Then, based on the prediction results, the load balancing and protection settings of the power grid are adjusted, such as initiating emergency load reduction or disconnecting certain high-risk connections, to evaluate the stability and resilience of the power grid when an actual fault occurs, and provide decision support for power grid operation and maintenance.

[0128] According to the grid stability impact assessment results, the signal processing parameters are adjusted to optimize the grid's responsiveness and fault tolerance to potential faults and output enhanced offset signals;

[0129] In the process of finally designing the enhanced offset signal, the nonlinear dynamic characteristics analysis of the fault signal and the impact assessment results on the grid stability were considered in detail; first, according to the fault type and impact assessment, the parameters of the signal processing algorithm were adjusted. The parameters included the signal amplification factor, the response factor weighting and the adjustment coefficient, which directly affected the sensitivity and effect of the signal adjustment; then, through simulation tests in a laboratory environment, the impact of the adjusted signal on the grid model was verified to ensure that the enhanced signal could effectively resist the unstable factors caused by potential faults; then, the enhanced signal was adjusted using real-time data feedback to adapt it to changes in the actual grid environment. Through continuous monitoring and adjustment, it was ensured that the signal adjustment could continue to adapt to the needs of grid operation. The design of the enhanced offset signal not only took into account the current grid conditions, but also reserved adjustment space to cope with possible changes in network load in the future.

[0130] See also Figure 7 , the steps to obtain the grid fault location results are:

[0131] Use the enhanced offset signal to perform reverse tracking to determine the propagation path and time delay of the traveling wave signal, and obtain the signal propagation path data by analyzing the speed and waveform characteristics of the traveling wave signal;

[0132] Using the acquired enhanced offset signal, the reverse tracking algorithm model is used to determine the propagation path and time delay of the traveling wave signal. First, real-time signal data is obtained from the data monitoring center of the entire network. These data include but are not limited to time series data of voltage, current and frequency. Each data point is recorded by a high-precision sensor at a specific time point. The original signal data is processed by a high-pass filter to filter out possible noise and interference. The discrete Fourier transform is further used to analyze the spectral characteristics of the signal and determine the main frequency components of the signal. Then, by comparing the time difference of the signal reaching different sensors and the known signal propagation speed, the propagation path and time delay of the traveling wave signal are calculated. These calculations are based on the wave speed and time delay formula. Through this method, the traveling wave signal in the power grid can be accurately tracked, and its propagation path can be determined, and finally the signal propagation path data can be generated.

[0133] Based on the signal propagation path data, combined with the grid topology and equipment location information, data fusion and analysis are performed to identify potential fault point locations and generate a list of candidate fault point locations;

[0134] The signal propagation path data is combined with the detailed topological structure of the power grid and the equipment location information, and the geographic information system (GIS) is used for data fusion and analysis. First, the signal propagation path data is imported into the GIS system. The system has been preloaded with a detailed map of the national power grid and the precise location information of the power grid equipment. The location information includes the specific coordinates of substations, transmission lines, and distribution networks. The path data and geographic location information are matched and analyzed through spatial data analysis methods to identify possible fault point locations on the signal propagation path. In this process, spatial analysis tools and models, such as buffer analysis and intersection detection, are used to identify the exact location where the fault may occur, and a list of possible fault points is generated, including the specific location of each candidate fault point and the corresponding signal characteristics.

[0135] Use the candidate fault location list to check with the actual geographic and equipment operation data, determine the exact location of the fault source through data comparison and analysis, and generate the power grid fault location result;

[0136] The candidate location list is used for detailed verification with the actual geographic and equipment operation data. First, the equipment operation data related to the fault is collected, including voltage, current, temperature and historical equipment fault records. The data is usually obtained from field sensors and historical databases. Each data point is subjected to time series analysis to determine the operation trend and possible abnormal patterns of the equipment. Then, the operation data is fused and analyzed with the geographic location data in the GIS system. Data matching and pattern recognition techniques are used to determine the correlation between the abnormal patterns in the data and the geographic location. Not only can the location of the fault point be identified, but also the cause and impact of the fault can be analyzed. Finally, the fault source location is accurately determined through algorithm optimization and data fusion technology, and the power grid fault location result is generated, which will directly support the maintenance team to carry out targeted repairs and troubleshooting.

[0137] A method for locating power grid faults based on traveling wave ranging comprises the following steps:

[0138] S1: Continuously monitor the traveling wave signal through the ADC device, continuously record the voltage value and corresponding acquisition time of each data point, identify each peak, and record the corresponding timestamp, calculate the total height value of the peak, and generate key morphological features;

[0139] S2: Based on the key morphological features, segment processing is performed in time sequence, change point detection is performed, the change coefficients of continuous signal points are calculated, each instantaneous change point is identified, the time coordinates of each change point are marked, and the corresponding relationship between the change point and the time series is generated;

[0140] S3: Based on the corresponding relationship between the change point and the time series, the change points in the time series are analyzed point by point, the fluctuation characteristics are numerically evaluated through the normalization method, the time period and amplitude change characteristics corresponding to the abnormal area are extracted and the increase and decrease trends are analyzed, the deviation direction of the overall waveform trend is determined, and the dynamic deviation prediction analysis results are generated;

[0141] S4: Using the dynamic offset prediction analysis results, extract the signal components with significant nonlinear characteristics, perform statistical analysis to characterize the dynamic change trend of the signal, and calculate and output the nonlinear dynamic index of the traveling wave signal;

[0142] S5: Based on the nonlinear dynamic index of the traveling wave signal, analyze the nonlinear dynamic characteristics of the fault signal, locate the fault area and type, evaluate the potential impact of the fault on the stability of the power grid, adjust the signal processing parameters, and output the enhanced offset signal;

[0143] S6: Use the enhanced offset signal to determine the propagation path and time delay of the traveling wave signal, analyze the speed and waveform characteristics of the traveling wave signal, combine it with the grid topology and equipment location information, identify the location of potential fault points, check it with the actual geographic and equipment operation data, and generate grid fault location results.

[0144] The above are only preferred embodiments of the present invention and are not intended to limit the present invention in other forms. Any technician familiar with the profession may use the technical contents disclosed above to change or modify them into equivalent embodiments with equivalent changes and apply them to other fields. However, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention without departing from the technical solution of the present invention still falls within the protection scope of the technical solution of the present invention.

Claims

1. A power grid fault location system based on traveling wave ranging, characterized in that: The system comprises: The signal capture and analysis module continuously collects traveling wave signals through ADC, measures the signal peak height, time interval and waveform change, timestamps and preliminarily classifies the data, identifies key events in the signal, and generates key morphological features; The dynamic tracking and identification module uses the key morphological features to track the signal waveform in detail, detect each instantaneous point of the traveling wave signal, analyze the correspondence between the signal change point and the time series, identify the traveling wave change pattern based on the corresponding relationship analysis results, evaluate the abnormal behavior of the traveling wave signal, analyze the change trend of the signal waveform, and generate dynamic offset prediction analysis results; The nonlinear characteristic analysis module extracts the nonlinear characteristic value in the traveling wave signal according to the dynamic offset prediction analysis result, calculates the nonlinear dynamic index of the traveling wave signal, analyzes the nonlinear dynamic characteristics of the fault signal according to the calculation result, evaluates the impact on the stability of the power grid, and outputs an enhanced offset signal; The fault location and reconstruction module reversely tracks the propagation path and time delay of the traveling wave signal based on the enhanced offset signal, analyzes the possible location of the fault point, determines the accurate location of the fault source by checking with the actual geographic data, and generates a power grid fault location result.

2. The power grid fault location system based on traveling wave ranging according to claim 1, characterized in that: The steps for obtaining the key morphological features are: The traveling wave signal is continuously monitored by the ADC device, and the voltage value and corresponding acquisition time of each data point are continuously recorded to generate a raw waveform data set with a time stamp; Based on the original waveform data set with time stamps, each peak is identified, and the corresponding timestamp is recorded, and the signal is preliminarily classified to obtain a classified peak feature data set; Based on the classified peak feature data set, the formula is adopted: Calculate the total peak height H and generate key morphological features, where p i represents the height of the ith peak, t i It represents the time interval between the ith peak and the previous peak, and n represents the total number of peaks.

3. The power grid fault location system based on traveling wave ranging according to claim 2, characterized in that: The analysis steps of the corresponding relationship are: Based on the key morphological features, segmentation processing is performed in time sequence, and the segmented waveform data is screened by numerical range to generate a signal waveform data set with time tags; The change point detection is performed on the signal waveform data set with time tags using the formula: Calculate the coefficient of variation C of consecutive signal points i , identify each instantaneous change point and obtain the instantaneous signal change point set, where S i and S i+1 is the value of two consecutive signal points; The instantaneous signal change point set is used to arrange the change points in sequence according to the time labels, mark the time coordinate of each change point, and generate a corresponding relationship result between the change point and the time series.

4. The power grid fault location system based on traveling wave ranging according to claim 3 is characterized in that: The steps for obtaining the dynamic shift prediction analysis result are as follows: Based on the corresponding relationship between the change points and the time series, the change points in the time series are analyzed point by point, and the time interval Δt between consecutive change points is calculated. i and amplitude difference ΔA i , and extract the fluctuation amplitude and fluctuation direction of each change point, summarize the change characteristics of continuous points, and generate a traveling wave change feature set; The traveling wave variation feature set is called, and the fluctuation feature is numerically evaluated by a normalization method, using the formula: Calculate the abnormal deviation value D of the overall fluctuation and generate the abnormal behavior result, where ΔA i represents the fluctuation amplitude of the ith point, is the average value of all fluctuation amplitudes, and N represents the total number of fluctuation points; Based on the abnormal behavior results and the traveling wave change feature set, the time period and amplitude change characteristics corresponding to the abnormal area are extracted segment by segment, the increase and decrease trends are analyzed, and the offset direction of the overall waveform trend is determined to generate dynamic offset prediction analysis results.

5. The power grid fault location system based on traveling wave ranging according to claim 4, characterized in that: The calculation steps of the nonlinear dynamic index are: Using the dynamic shift prediction analysis results, deep spectrum analysis is performed to extract signal components with significant nonlinear characteristics and generate a traveling wave signal nonlinear characteristic value sequence; Based on the nonlinear characteristic value sequence of the traveling wave signal, statistical analysis is performed to determine the mean, standard deviation and kurtosis coefficient, characterize the dynamic change trend of the signal, and generate statistical dynamic parameters of the traveling wave signal; Combining the nonlinear characteristic value sequence of the traveling wave signal and the statistical dynamic parameters of the traveling wave signal, the formula is adopted: The nonlinear dynamic index NDI of the traveling wave signal is calculated, where f i represents the i-th nonlinear eigenvalue in the traveling wave signal, p i Represents the corresponding statistical dynamic parameter, μ represents the mean of the nonlinear eigenvalue sequence, and m represents the total number of nonlinear eigenvalues.

6. The power grid fault location system based on traveling wave ranging according to claim 5, characterized in that: The steps of acquiring the enhanced offset signal are: Based on the nonlinear dynamic index of the traveling wave signal, the nonlinear dynamic characteristics of the fault signal are analyzed, the potential fault mode matching the current index is identified, the fault area and type are located, and the nonlinear dynamic characteristics analysis result of the fault signal is generated; Using the nonlinear dynamic characteristic analysis results of the fault signal, simulating the power grid response according to the severity of the fault characteristics, evaluating the potential impact of the fault on the power grid stability, and generating a power grid stability impact assessment result; According to the grid stability impact assessment result, the signal processing parameters are adjusted to optimize the grid's responsiveness and fault tolerance to potential faults, and an enhanced offset signal is output.

7. The power grid fault location system based on traveling wave ranging according to claim 6, characterized in that: The steps for obtaining the power grid fault location result are: Using the enhanced offset signal, reverse tracking is performed to determine the propagation path and time delay of the traveling wave signal, and by analyzing the speed and waveform characteristics of the traveling wave signal, signal propagation path data is obtained; Based on the signal propagation path data, combined with the power grid topology and equipment location information, data fusion and analysis are performed to identify potential fault point locations and generate a list of candidate fault point locations; The candidate fault location list is used to check with actual geographic and equipment operation data, and the exact location of the fault source is determined through data comparison and analysis to generate a power grid fault location result.

8. A method for locating power grid faults based on traveling wave ranging, characterized in that: The system according to any one of claims 1 to 7 is implemented, comprising the following steps: The ADC device continuously monitors the traveling wave signal, continuously records the voltage value and corresponding acquisition time of each data point, identifies each peak, records the corresponding timestamp, calculates the total height value of the peak, and generates key morphological features; Based on the key morphological features, segment processing is performed in time sequence, change point detection is performed, the change coefficients of continuous signal points are calculated, each instantaneous change point is identified, the time coordinates of each change point are marked, and the corresponding relationship between the change point and the time series is generated; Based on the corresponding relationship between the change points and the time series, the change points in the time series are analyzed point by point, the fluctuation characteristics are numerically evaluated by the normalization method, the time period and amplitude change characteristics corresponding to the abnormal area are extracted and the increase and decrease trends are analyzed, the deviation direction of the overall waveform trend is determined, and the dynamic deviation prediction analysis results are generated; Utilizing the dynamic offset prediction and analysis results, extracting the signal components with significant nonlinear characteristics, performing statistical analysis to characterize the dynamic change trend of the signal, and calculating and outputting the nonlinear dynamic index of the traveling wave signal; Based on the nonlinear dynamic index of the traveling wave signal, the nonlinear dynamic characteristics of the fault signal are analyzed, the fault area and type are located, the potential impact of the fault on the stability of the power grid is evaluated, the signal processing parameters are adjusted, and the enhanced offset signal is output; The enhanced offset signal is used to determine the propagation path and time delay of the traveling wave signal, analyze the speed and waveform characteristics of the traveling wave signal, combine it with the power grid topology and equipment location information, identify the location of the potential fault point, and check it with the actual geographical and equipment operation data to generate a power grid fault location result.

Images