Alternating current and direct current integrated grounding fault positioning method and system

Abstract

The invention relates to an AC-DC integrated grounding fault positioning method and system, and the method comprises the steps: collecting the potential change data of a DC power supply system and the frequency characteristic data of an AC power supply system, and determining the electrical signal characteristics of stray current corrosion; decomposing the electric signal by using wavelet transform, and extracting direct-current and alternating-current components; determining a fault type according to the component amplitude; analyzing the potential gradient and the stray current flow direction in combination with a line topological structure, and determining the position of a fault point; measuring resistance to determine a grounding loop, and analyzing the correlation between corrosion distribution and electrical parameter characteristics; evaluating the influence of corrosion on the grounding path in combination with the material resistivity; according to the method, faults can be accurately recognized, fault points can be accurately positioned, the corrosion degree and the influence of the corrosion degree on the system can be evaluated, and effective support is provided for safe operation and maintenance of a subway power supply system.

Description

Technical Field

[0001] This application relates to the technical field of measuring electrical variables, and specifically relates to an AC-DC integrated grounding fault location method and system. Background Art

[0002] The subway traction power supply system uses direct current, while the auxiliary system relies on alternating current. This coexistence of AC and DC is particularly prominent in a complex electromagnetic environment. However, the stray current caused by grounding faults will corrode metal pipes and equipment casings, threatening the safety of train operation. Existing methods mostly rely on traditional electrical detection technologies, such as voltage monitoring or current distribution analysis, but it is difficult to accurately distinguish the fault source under strong electromagnetic interference. Especially in an AC-DC hybrid system, the fault characteristics are easily masked by interference signals, resulting in insufficient positioning accuracy. Existing technologies are difficult to effectively separate the fault characteristics of the traction network and the signal system in a strong electromagnetic interference environment, leading to low fault diagnosis efficiency and reduced system reliability. Therefore, how to utilize the AC-DC characteristics of the subway power supply system to accurately distinguish the grounding faults of the traction network and the signal system and achieve precise positioning of the fault point in a strong electromagnetic interference environment has become a key issue in improving system safety and operation stability. Summary of the Invention

[0003] In order to solve the problems existing in the above-mentioned prior art, the purpose of this application is to provide an AC-DC integrated grounding fault location method and system.

[0004] An AC-DC integrated grounding fault location method described in this application includes the following steps: Collect the potential change data of the subway traction DC power supply system and the frequency characteristic data of the AC power supply system to determine the electrical signal characteristics of stray current corrosion; Under a strong electromagnetic interference environment, perform 4-layer wavelet decomposition on the electrical signal using the db4 wavelet basis function, and extract the AC component eigenvalue in the high-frequency component and the DC component eigenvalue in the low-frequency component; Calculate the amplitude of the DC component eigenvalue. If it exceeds the preset DC amplitude threshold, it is determined as a DC grounding fault. Calculate the amplitude, waveform distortion degree, and phase shift angle of the AC component eigenvalue. If the amplitude exceeds the preset AC amplitude threshold and the waveform distortion degree > 3% or the phase shift angle > 30°, it is determined as an AC grounding fault; Combined with the subway line topology structure, iteratively calculate the grid node potential values based on the finite element method, generate a potential gradient distribution map, calculate the current density between nodes through the current continuity equation, and determine the stray current flow vector in combination with the potential gradient direction; Based on the potential gradient distribution map and the stray current flow vector, determine the physical location interval of the fault point. If it is determined that there is a physical location interval of the fault point, then execute: measure the resistance between the pipeline or equipment shell and the preset grounding point, extract the corrosion distribution data by combining the potential gradient distribution and the stray current flow vector, obtain the material resistivity, analyze the current density distribution and potential gradient change in the corrosion area, and evaluate the degree of local resistance increase, impedance discontinuity, and signal attenuation; Send a narrow pulse signal with a pulse width ≤ 1 μs to the traction network and the signal system, calculate the initial fault distance according to the propagation delay and amplitude attenuation data of the reflected waveform, and correct the waveform interference offset based on the evaluation results of the degree of local resistance increase, impedance discontinuity, and signal attenuation, and output the calibrated fault point coordinates.

[0005] An AC-DC integrated grounding fault location system described in this application includes: A potential acquisition module, which is used to acquire the potential change in the subway traction DC power supply system and the frequency characteristics in the AC power supply system, and determine the electrical signal characteristics of the occurrence of stray current corrosion; A signal decomposition module, which is used to perform 4-layer wavelet decomposition on the electrical signal using the db4 wavelet basis function in a strong electromagnetic interference environment, and extract the AC component eigenvalue in the high-frequency component and the DC component eigenvalue in the low-frequency component; A fault determination module, which is used to calculate the amplitude of the DC component eigenvalue. If it exceeds the preset DC amplitude threshold, it determines a DC grounding fault. It calculates the amplitude, waveform distortion degree, and phase shift angle of the AC component eigenvalue. If the amplitude exceeds the preset AC amplitude threshold and the waveform distortion degree > 3% or the phase shift angle > 30°, it determines an AC grounding fault; A fault location module, which is used to combine the subway line topology structure, iteratively calculate the grid node potential values based on the finite element method, generate a potential gradient distribution map, calculate the current density between nodes through the current continuity equation, and determine the stray current flow vector by combining the potential gradient direction; A corrosion analysis module, which is used to determine the physical location interval of the fault point based on the potential gradient distribution map and the stray current flow vector. If it is determined that there is a physical location interval of the fault point, then execute: measure the resistance between the pipeline or equipment shell and the preset grounding point, extract the corrosion distribution data by combining the potential gradient distribution and the stray current flow vector, obtain the material resistivity, analyze the current density distribution and potential gradient change in the corrosion area, and evaluate the degree of local resistance increase, impedance discontinuity, and signal attenuation; A fault calibration module, which is used to send a narrow pulse signal with a pulse width ≤ 1 μs to the traction network and the signal system, calculate the initial fault distance according to the propagation delay and amplitude attenuation data of the reflected waveform, and correct the waveform interference offset based on the evaluation results of the degree of local resistance increase, impedance discontinuity, and signal attenuation, and output the calibrated fault point coordinates.

[0006] A method and system for integrated AC-DC grounding fault location according to the present application has the advantages that by collecting the potential and frequency characteristics of the subway traction power supply system, decomposing the electrical signal using wavelet transform, extracting DC and AC components, judging the fault type according to the component amplitude, analyzing the potential gradient and the flow direction of stray current in combination with the line topology structure to determine the fault point location, determining the grounding loop by measuring the resistance, analyzing the correlation between the corrosion distribution and the electrical parameter characteristics, evaluating the influence of corrosion on the grounding path in combination with the material resistivity, sending a pulse signal and analyzing the reflection waveform to determine the fault distance. The present invention can accurately identify the stray current corrosion fault, locate the fault point, evaluate the corrosion degree and its influence on the system, and provide effective support for the safe operation and maintenance of the subway power supply system. BRIEF DESCRIPTION OF THE DRAWINGS

[0007] Figure 1 is the flow of a method for integrated AC-DC grounding fault location according to the present application Figure 1 。 DETAILED DESCRIPTION OF THE EMBODIMENTS

[0008] As Figure 1 shown, a method for integrated AC-DC grounding fault location according to the present application includes the following steps: As Figure 1 shown, in step S101, potential change data is collected in the subway traction DC power supply system and frequency characteristic data is collected in the AC power supply system to determine the electrical signal characteristics caused by stray current corrosion, where the electrical signal characteristics include the DC potential offset amplitude and duration, and the corrosion occurrence interval is judged.

[0009] Specifically, in step S101, a distributed data acquisition unit is used to collect original DC potential data from the track foundation and the grounding grid nodes of the subway traction DC power supply system in real time, the sampling interval is set to 100 milliseconds, and the potential fluctuation value is recorded according to the preset DC potential sampling frequency; According to the monitoring requirements of the substation, the acquisition duration and interval are set, potential sampling is performed on the spatial distribution positions of different measurement points, and a DC potential data cache queue is established; The original DC potential data is subjected to 4-layer wavelet decomposition to remove high-frequency noise, and then a median filter window with a length of 5 is used to eliminate sudden interference, and a smooth preprocessed DC potential data sequence is obtained. For example, the original potential of a certain measurement point is 850 mV, and it stabilizes at 820 mV after processing; In the AC power supply system, a signal analyzer is used to collect the voltage signal according to the AC frequency sampling period, perform 1024-point fast Fourier transform in the frequency range of 0 to 1000 Hz to generate an AC frequency characteristic spectrum, and apply a Hanning window function for smoothing processing to reduce spectrum leakage; During the operation of the substation, the frequency characteristic spectrum shows the fundamental frequency of 50 Hz and its harmonic components, where the amplitude of the third harmonic accounts for 15% of the fundamental wave, and the fifth harmonic accounts for 8% of the fundamental wave; through time series correlation analysis, the preprocessed DC potential data sequence is matched with the AC frequency characteristic spectrum to extract the characteristics of the electrical signal; Perform an offset amplitude detection on the preprocessed DC potential data sequence, set the positive threshold to 600 mV and the negative threshold to 1100 mV. When the potential offset exceeds the threshold and lasts for more than 300 seconds, trigger the calculation of stray current; Adopt the sliding time window method. Based on the track loop resistance of 0.01 ohm, the amplitude of the stray current is obtained by dividing the potential difference by the resistance. For example, when the potential difference is 200 mV, the stray current is 20 A, and the amplitude sequence is recorded; Further construct a three-layer BP neural network mapping model. The input layer contains 12 nodes, including the mean value, standard deviation, peak factor of the stray current, and the characteristics of the AC harmonic content. The hidden layer has 16 nodes, and the output layer is the determination value of the potential anomaly degree. The sigmoid activation function is used to train the model; In practical applications, compare the output value of the neural network with the preset anomaly threshold of 0.85. If the determination value exceeds the standard, it is confirmed that stray current corrosion has occurred. For example, in a certain detection, the potential offset lasts for 875 seconds, and the maximum amplitude reaches 350 mV, which appears simultaneously with the increase of the AC third harmonic, indicating the existence of corrosion risk; Record the characteristics of the potential offset amplitude and duration within the corrosion interval to provide data support for subsequent fault location; the embodiments of the present invention do not limit the specific sampling frequency or filtering parameters too much, and can be adjusted by technicians according to the actual scenario; Through distributed acquisition and signal processing technology, the characteristics of electrical signals in the AC-DC hybrid environment can be effectively extracted. The neural network analysis further improves the recognition accuracy of the corrosion interval, laying a foundation for the judgment of fault types. Compared with traditional methods, this method shows higher robustness in a strong interference environment, ensuring the reliability of fault diagnosis; In practical applications, the corrosion interval is triggered when the determination value of the potential anomaly degree output by the neural network exceeds the preset threshold.

[0010] In one embodiment, the calculation of the stray current is that the potential difference refers to the measured potential difference between the track and the grounding grid, and the resistance refers to the preset standard measurement resistance (0.01 Ω).

[0011] In one embodiment, the corrosion interval is defined as the time period when the potential offset lasts for more than 300 seconds.

[0012] Such as Figure 1As shown, in step S102, the characteristics of the electrical signal are decomposed by using wavelet transform technology in a strong electromagnetic interference environment to obtain high-frequency components and low-frequency components, and the DC components of the DC power supply system and the AC components of the AC power supply system are extracted therefrom.

[0013] Specifically, in step S102, in a strong electromagnetic interference scenario, a high-precision data collector samples the electrical signal according to the time sequence, and the environmental interference intensity is monitored in real time through an electromagnetic field intensity sensor, and the sampling frequency is adaptively adjusted according to the interference intensity to obtain the original sampling signal; When the interference intensity exceeds 100 microteslas, the sampling frequency is automatically increased to 40 kHz to ensure the capture of high-frequency interference details; The original sampling signal contains AC-DC mixed characteristics and noise interference. To improve the signal quality, a band-pass filter is used to set the passband range from 5 Hz to 5 kHz to filter out low-frequency power frequency interference and high-frequency random noise. At the same time, the pulse detection circuit is used to identify the mutation interference points. For example, after detecting a pulse interference with an amplitude of 2 V and a duration of 50 μs, the signal cancellation algorithm is used to replace the interference section with the mean value of adjacent sampling points to generate a preprocessed sampling signal; The preprocessed sampling signal is subjected to wavelet transform processing. The Daubechies 4 wavelet basis function is selected because of its superior time-frequency resolution ability. The signal is decomposed into multiple frequency band sub-signals through four-layer decomposition. Among them, the first-layer decomposition extracts the high-frequency sub-signal from 2 to 5 kHz, the second layer extracts the frequency band signal from 1 to 2 kHz, the third layer extracts the frequency band signal from 500 Hz to 1 kHz, and the fourth layer separates the frequency band signal from 250 to 500 Hz and the low-frequency sub-signal below 250 Hz; To ensure the frequency band separation effect, calculate the energy ratio of each frequency band. For example, a certain measurement shows that the energy of the high-frequency sub-signal accounts for 35% and the low-frequency sub-signal accounts for 65%. According to this frequency band separation degree, set the high-frequency threshold of 0.5 V and the low-frequency threshold of 0.8 V, and perform soft threshold processing on the high-frequency and low-frequency sub-signals respectively, retain the effective components greater than the threshold, and remove the weak noise to obtain the filtered high-frequency component and low-frequency component; Extract the characteristics of the AC power supply system from the filtered high-frequency component, and identify the fundamental frequency of 50 Hz and its harmonic components through spectrum analysis. For example, the amplitude of 100 Hz is 0.3 V, the amplitude of 150 Hz is 0.2 V, and the amplitude of 200 Hz is 0.1 V. Extract the characteristics of the DC power supply system from the filtered low-frequency component, and detect that the DC voltage value is 750 V and is accompanied by low-frequency fluctuations below 20 Hz; Integrate the extracted characteristic values into a 16-dimensional feature vector, including 4 AC frequency parameters, 4 phase parameters, 4 amplitude parameters, and 4 DC parameters; The support vector machine is used to classify and train the feature vectors with a radial basis kernel function. Under normal conditions, the feature vectors are clustered and distributed in the high-dimensional space. When an anomaly occurs, they deviate from the clustering region. For example, when the DC voltage fluctuation exceeds 50 volts and the amplitude of the 150 Hz harmonic suddenly increases to 0.4 volts, it is judged as an abnormal state, and the discrimination result of the electrical signal features is output. For the signal decomposition requirements in a strong electromagnetic interference environment, wavelet transform effectively separates high and low frequency components through multi-scale analysis. Compared with the traditional Fourier transform, it is more suitable for processing non-stationary signals. The combination of band-pass filtering and signal cancellation algorithm ensures the purity of the preprocessed signal, providing a reliable basis for subsequent feature extraction. The classification and discrimination of the support vector machine map the signal characteristics under complex interference into recognizable patterns through high-dimensional feature space mapping, improving the discrimination accuracy. In practical applications, if the electromagnetic interference intensity fluctuates frequently, the data collector can dynamically adjust the sampling interval according to the interference intensity. For example, when the interference intensity rises to 150 microteslas, the sampling frequency is further increased to 50 kHz. During the wavelet decomposition process, an adaptive threshold adjustment mechanism can be introduced for the calculation of the frequency band separation degree, and the high and low frequency threshold settings are optimized according to the real-time energy ratio to ensure the accuracy of the decomposed components. The construction of the feature vectors can also be extended to 20 dimensions, and new time-domain statistical features such as the peak factor and skewness are added to further enhance the classification robustness. In some scenarios, if it is detected that the harmonics in the high-frequency components are abnormally prominent, for example, the amplitude of 150 Hz exceeds 0.4 volts and the duration exceeds 1 second, an abnormal warning for the AC component can be directly triggered. If the DC voltage fluctuation amplitude in the low-frequency component exceeds 50 volts and is accompanied by low-frequency oscillation, it can be preliminarily judged as an abnormal DC component. When training the support vector machine, a multi-class classification model can be introduced to distinguish the four states of normal state, DC anomaly, AC anomaly, and mixed anomaly, and output a more refined discrimination result to guide subsequent fault analysis. Through adaptive sampling and multi-level signal processing, it is possible to accurately extract the AC and DC component features in a complex electromagnetic environment. The application of the support vector machine not only improves the discrimination efficiency but also provides data support for subsequent determination of fault types. In actual operation, this method can flexibly adjust parameters according to environmental changes to ensure applicability under different interference intensities.

[0014] In one embodiment, the electrical signal sampling object is to sample the rail potential signal of the subway traction DC power supply system and the voltage signal of the AC power supply system.

[0015] In one embodiment, the support vector machine classifier uses a radial basis kernel function (RBF), with a penalty factor C = 1.0 and a kernel parameter γ = 0.01. The training data consists of 200 sets of historical fault samples (100 sets in normal state, 50 sets of DC faults, and 50 sets of AC faults).

[0016] As Figure 1 shown, in step S103, the amplitudes of the DC component and the AC component are calculated respectively and compared with a preset threshold to determine whether there is a DC or AC grounding fault and identify the fault type.

[0017] Specifically, in step S103, a high-precision data acquisition device is used to sample the grounding point voltage signal. The sampling period is set to 100 microseconds. The original voltage signal is processed by an 8th-order Butterworth digital filter to eliminate high-frequency interference components above 10 kHz. Subsequently, a 1024-point fast Fourier transform is performed on the filtered signal for spectral analysis to obtain the DC voltage amplitude and the AC voltage amplitude respectively, providing a data basis for subsequent fault determination. In the processing of the DC component, the DC voltage amplitude is smoothed using a mean sliding window with a length of 10 to reduce the influence of instantaneous fluctuations. Considering the drift characteristics of the measuring device at different ambient temperatures, an amplitude correction coefficient is introduced for error compensation. For example, the correction coefficient is set to 1.02 at an ambient temperature of 35 degrees Celsius. The preset DC voltage threshold range is from 1200 volts to 1800 volts. If the corrected DC voltage amplitude exceeds this range, for example, it drops to 1150 volts in a certain measurement, the DC voltage out-of-limit flag is recorded and the out-of-limit timestamp is marked. This method ensures the stability and accuracy of the DC amplitude calculation through double processing of smoothing and compensation, avoiding misjudgment caused by noise or device errors. In the analysis of the AC component, the fundamental 50 Hz component is extracted through spectral analysis and the AC voltage amplitude is calculated. At the same time, the waveform distortion degree and the phase shift angle are calculated to characterize the abnormal features of the signal. During normal operation, the distortion degree is below 3% and the phase shift angle is close to zero. The preset AC voltage threshold is 10 volts. If the measured amplitude exceeds this value, for example, it reaches 12 volts and the phase shift angle exceeds 30 degrees, the AC voltage out-of-limit flag is recorded. This process can effectively identify abnormal fluctuations in the AC signal through multi-dimensional feature extraction, providing rich information for subsequent fault type determination. The measured value of the grounding resistance is obtained using a grounding resistance measuring device based on the four-wire method principle. The measuring current is set to 10 amperes. The influence of ambient temperature and contact resistance is eliminated through a resistance measurement compensation algorithm. For example, the normal resistance value should be less than 0.1 ohm. If it rises to 0.3 ohm in a certain measurement, it indicates an abnormal grounding path. Perform a sequential correlation analysis on the DC voltage over - limit flag, AC voltage over - limit flag, and the corrected ground resistance value, and calculate the over - limit duration. For example, if both exceed the limit and last for more than 1 second during a certain detection, it indicates significant fault characteristics. In the fault type discrimination, use the Gaussian mixture clustering algorithm to cluster the feature vectors including the duration. The feature vectors include the DC voltage over - limit flag, AC voltage over - limit flag, and over - limit duration, generating three types of fault feature distributions, including normal state, minor anomaly, and severe fault. Further establish a fault discrimination model through the random forest algorithm. The model contains 50 decision trees, and the input features include the DC voltage deviation value, AC voltage amplitude, waveform distortion degree, phase shift angle, and ground resistance value. The random forest maps the fault feature relationship through a multi - tree voting mechanism. For example, during a certain operation, the DC voltage drops to 1150 V, the AC voltage rises to 12 V, and the ground resistance value is 0.3 ohms. The model output result indicates a double - ground fault, with the DC fault located at the negative busbar and the AC fault involving the A - phase winding. The combination of Butterworth filter and Fourier transform ensures the accuracy of signal processing. The multi - step processing of amplitude calculation and threshold determination improves the reliability of fault detection. The real - time measurement of the ground resistance value provides an auxiliary basis for fault location, and the combined application of Gaussian mixture clustering and random forest algorithm realizes the accurate identification of complex fault types through feature clustering and classification prediction. In practical applications, if the DC voltage amplitude fluctuates frequently, the sliding window length can be dynamically adjusted. For example, it can be increased to 15 sampling points in a high - interference environment to enhance the smoothing effect. For AC signals, if an abnormal increase in harmonic components is detected, additional features such as the harmonic content ratio can be introduced to further refine the distortion degree analysis. The fault discrimination model can also regularly update the decision tree weights according to the operation data to adapt to the long - term change characteristics of the subway power supply system, thereby improving the discrimination robustness. This method can quickly distinguish fault types in the AC - DC hybrid power supply system through multi - parameter comprehensive analysis. Compared with traditional single - threshold judgment, the random forest model uses multi - dimensional feature mapping to significantly improve the discrimination accuracy, especially suitable for complex ground fault scenarios.

[0018] In one embodiment, calculate the waveform distortion degree and phase shift angle for the AC voltage signal, where: The waveform distortion degree is calculated by extracting the fundamental wave and harmonic components through the fast Fourier transform; the phase shift angle is obtained by comparing the reference waveform through the cross - correlation algorithm. If the amplitude of the AC component exceeds the preset AC amplitude threshold, and the waveform distortion degree > 3% or the phase shift angle > 30°, then an AC grounding fault is determined, which is used to clarify that the calculation object is the AC voltage signal rather than the amplitude.

[0019] In one embodiment, the determination logic of the AC overlimit flag is that the determination of the AC overlimit flag needs to simultaneously meet: the amplitude of the AC component > 10V, and the waveform distortion degree > 3% or the phase shift > 30°.

[0020] In one embodiment, the resistance value example logic is that after eliminating the influence of the ambient temperature through the compensation algorithm, if the grounding resistance value > 0.1Ω (for example, the measured compensated value is 0.3Ω), then the grounding path is determined to be abnormal.

[0021] In one embodiment, the random forest fault discrimination model: the number of decision trees is 50, the feature splitting rule uses the Gini impurity, and the input feature weight distribution is the DC voltage deviation (weight 0.3), the AC voltage amplitude (0.25), the waveform distortion degree (0.2), the phase shift angle (0.15), and the grounding resistance value (0.1).

[0022] As Figure 1 shown, in step S104, by combining the fault type with the subway line topology structure, the potential gradient distribution and the characteristics of the stray current flow direction are analyzed to determine the physical location interval of the fault point.

[0023] Specifically, in step S104, the line connection relationship and the section division information are extracted based on the subway line topology diagram, and the line section is spatially modeled using the regular hexagon grid division method. The side length of each grid is set to 0.5 meters, covering the entire line length. For example, a 25-kilometer line is divided into approximately 50,000 basic grid units; Combined with the three-dimensional coordinates of the grounding point, for example, a grounding point is measured to be 116.123 degrees east longitude, 39.456 degrees north latitude, and 45 meters above sea level through the global positioning system, and the initial section potential distribution data is established, laying a foundation for the subsequent potential field analysis; Based on the grid modeling, the initial section potential distribution data is refined. For the complex sections of the station and the tunnel, the triangular unit is used to encrypt the grid. For example, the number of grids in a certain station area increases from the basic 50,000 to 80,000; The potential field distribution equation is constructed by the finite element method, considering the material properties of the rail conductivity of 5.8×10^7 Siemens / meter and the concrete conductivity of 1×10^-6 Siemens / meter, setting the convergence threshold of 0.001 volts and the maximum number of iterations of 1000 times, and iteratively calculating the potential values of each grid node to generate the section potential field distribution data; For example, the potential of a certain rail node is measured to be 850 millivolts, and the adjacent concrete node is 50 millivolts, reflecting a significant potential difference; Based on the sectional potential field distribution data, the central difference method is used to calculate the potential gradient between adjacent grid nodes. For example, the potential gradient between the rail and the concrete reaches 1.6 volts per meter; To extract spatial features, an 8-layer convolutional neural network is used to process the potential gradient distribution map. The input size is 120×120 pixels, and the convolutional kernel size is 3×3. By means of multi-layer convolution and pooling operations, the local gradient change law is captured to generate the sectional potential gradient distribution feature map; Compared with the traditional differential calculation, this method can more comprehensively characterize the spatial distribution characteristics of the potential field and provide high-dimensional feature support for fault location; In current analysis, the current continuity equation is used to calculate the current density between grid nodes, and the current value is corrected by combining the cross-sectional area of the rail, which is 7700 square millimeters, and the material conductivity. For example, the current density at a certain place is calculated as 0.12 amperes per square centimeter, and the direction points to the concrete structure; Spatial clustering is performed on the current flow vector to identify the current inflow and outflow regions. For example, the cumulative current inflow in the station platform area is 15 amperes, and the outflow in the adjacent track section is 12 amperes, revealing the unbalanced distribution of stray current; The integral value of the current density in each region is quantified through the clustering results to generate the current inflow and outflow distribution characteristics of the structure; The potential gradient distribution characteristics and the current flow characteristics are compared with the pre-calibrated fault feature library, and a deep regression tree is used to predict the spatial coordinates of the fault point; The training samples include 50 groups of typical fault data, each group including potential gradient, current density and flow vector, and the prediction error is controlled within 1.5 meters; For example, in a certain analysis, the predicted fault point is located at K15+750 of Line 3. The on-site verification shows that the potential of the concrete structure at this place is extremely low, reaching 300 millivolts, which is consistent with the model output, indicating that the fault point is closely related to the leakage of stray current; Spatial grid modeling converts complex lines into computable units. The finite element method accurately describes the potential field distribution through numerical solution, while the convolutional neural network enhances the depth and breadth of feature extraction. Compared with traditional manual analysis, this method significantly improves the positioning efficiency, especially in complex structural areas such as stations, and can effectively reveal the convergence law of stray current; In practical applications, if the line section involves the junction of multiple materials, the grid density can be dynamically adjusted. For example, the grid side length at the junction of the rail and the concrete is reduced to 0.2 meters; The temperature correction factor can be introduced into the current density calculation to adapt to environmental changes; The deep regression tree model can also regularly update the training samples and incorporate new fault modes to ensure that the prediction accuracy is optimized with the running time; Through multi-level analysis, this method not only locates the fault points but also reveals the influence range of stray current on the surrounding structures. Due to the complex grounding conditions in the station area, it often becomes a hot spot for current convergence, and it is necessary to strengthen monitoring to reduce the corrosion risk. During actual operation, the grid division and algorithm parameters can be flexibly adjusted according to the line characteristics to improve the applicability of the method. In practical applications, the method for obtaining the flow characteristics of stray current: Calculate the current density between nodes through the current continuity equation, and determine the stray current flow vector in combination with the direction of the potential gradient.

[0024] In one embodiment, the source of the potential distribution data is the initial section potential distribution data generated by iterative finite element method based on the measured values of the track potential sensors. In one of the embodiments, the definition method of the initial section can be: natural segmentation based on the distance between stations in the subway line topology.

[0025] In one embodiment, the finite element iteration parameters (convergence threshold 0.001V, maximum iteration 1000 times): Based on the convergence test results under the condition of the rail conductivity of 5.8×10 7 S / m, when the threshold <0.001V, the positioning error <0.1%; Dynamic adjustment formula for high and low frequency thresholds: High frequency threshold = 0.2×(electromagnetic interference intensity / 100μT)+0.3V; Low frequency threshold = 0.3×(band separation degree)+0.5V.

[0026] As Figure 1 shown, in step S105, if it is determined in step S104 that there is a physical location interval of the fault point, then step S105 is executed. Determine the grounding loop path by measuring the resistance between the pipeline and the equipment shell and the preset grounding point, and extract the corrosion distribution data in combination with the potential gradient distribution and the flow characteristics of the stray current, and analyze the correlation characteristics between the corrosion of the pipeline and the equipment shell and the fault point to reveal the fault influence mechanism.

[0027] Specifically, in step S105, arrange measurement electrodes based on the coordinates of the preset grounding point, apply a 1 ampere test current using the four-terminal method, and calculate the loop resistance by detecting the voltage drop between the pipeline and the grounding point through the voltage electrode. For example, the measured voltage drop at a certain place is 0.15 volts, and the resistance value is 0.15 ohms; To eliminate the influence of temperature, introduce a temperature compensation coefficient. For example, when the measured resistance is 0.156 ohms in an environment of 35 degrees Celsius, multiply it by 0.96 and correct it to 0.15 ohms; Construct loop resistance distribution data through multi-point measurement to reflect the electrical characteristics of the grounding loop and provide basic parameters for subsequent corrosion analysis; In corrosion detection, a 5-MHz ultrasonic flaw detector was used to conduct a grid scan of the pipeline surface. The scanning interval was set at 50 mm, covering 200 measurement points. Multiple corrosion pits were found in a certain section of the pipeline, with a depth range of 1 to 2.8 mm and a diameter of 15 to 40 mm; At the same time, an infrared thermal imager was used to collect the surface temperature distribution of the equipment shell. It was detected that the temperature in the corrosion area was 0.8 to 1.2 degrees Celsius lower than that of the surrounding area, and the area of the abnormal area was approximately 400 square centimeters; These initial data were subjected to spatial grid division. A continuous corrosion depth distribution surface was generated by cubic spline interpolation with an interpolation interval of 10 mm. Then, the elliptical corrosion boundary was fitted by the least squares method, with a major axis of 80 cm and a minor axis of 50 cm, accurately characterizing the spatial morphology of the corrosion; Combined with the characteristics of the potential gradient distribution, a potential gradient threshold of 0.5 V / m was set to divide the potential abnormal area. When the potential gradient exceeded 0.5 V / m, it was classified as an abnormal area. For example, the area of a certain abnormal area reached 600 square centimeters, and the overlap with the corrosion distribution data reached 75%, indicating a significant spatial correspondence between corrosion and potential abnormality; The feature vectors of the corrosion location were extracted through a deep neural network. The input included the depth, area of the corrosion pits, and the coordinates of the temperature abnormal points. After multi-layer convolution processing, a high-dimensional feature representation was generated for correlation analysis; In the assessment of the corrosion severity, a three-level judgment criterion was established: mild corrosion (average depth < 1 mm, area ratio < 5%), moderate corrosion (average depth 1 - 2 mm, area ratio 5 - 10%), and severe corrosion (average depth > 2 mm, area ratio > 10%). In this example, the average corrosion depth was 1.5 mm, the maximum depth was 2.8 mm, and the corrosion area ratio was 12%, so it was judged as severe corrosion; The corresponding electrical parameter characteristics included a loop resistance of 0.15 ohms, a potential difference of 350 mV, and a current density of 0.08 A / cm²; A mapping model between the corrosion location feature vectors and the electrical parameter characteristics was constructed using support vector regression. The radial basis kernel function was used in the training process. It was found that when the potential difference exceeded 300 mV and the current density was greater than 0.05 A / cm², the growth rate of the corrosion depth increased significantly, revealing the driving effect of electrical abnormality on corrosion; According to the central coordinates of the corrosion area, for example, located at K15 + 800 of the subway line, with a spatial distance of 15 m from the fault point, combined with the analysis of the stray current flow characteristics of the grounding loop path, it was found that the stray current at the fault point was conducted to this section of the pipeline through the concrete return channel, resulting in aggravated local corrosion; In one embodiment, the mapping relationship between the quantification index and the electrical parameter characteristics is quantified through support vector regression analysis, and the correlation coefficient is calculated to be 0.85, indicating a strong correlation between the corrosion degree and the change of electrical parameters. The association characteristic data between the corrosion distribution data and the fault point is generated, reflecting the dual characteristics of location proximity and electrical parameter characteristic coupling, providing a basis for tracing the cause of the fault; In practical applications, if the pipeline material or environmental humidity changes, the ultrasonic probe frequency can be adjusted to 7 MHz to improve the detection accuracy; The support vector regression model can introduce more features, such as the resistivity of the pipeline material and the humidity factor, to further optimize the mapping accuracy; Through long-term monitoring, it is found that under the continuous action of stray current, the boundary of the corrosion area may expand over time, and the measurement data needs to be updated regularly to track the dynamic changes; This method not only accurately locates the grounding loop through multi-dimensional detection means and data fusion analysis, but also deeply reveals the causal relationship between corrosion and the fault point. In actual operation, the aggravation of pipeline corrosion is often highly correlated with the leakage path of stray current, and it is necessary to strengthen protection in combination with the line maintenance strategy.

[0028] As Figure 1 shown, in step S106, the resistivity of the pipeline and equipment shell materials is obtained, and combined with the association characteristics between corrosion and the fault point, the current density distribution and potential gradient change of stray current in different materials and corrosion areas are analyzed, the law of corrosion distribution data is deduced, and its impact on the electrical characteristics of the grounding path is evaluated, including local resistance increase, impedance discontinuity, and signal attenuation aggravation.

[0029] Specifically, in step S106, the four-electrode method is used to measure the resistivity of the pipeline and shell materials, the electrode spacing is set to 10 cm, and the initial resistivity is calculated by injecting 1 A measurement current and detecting the voltage drop. For example, the measured resistivity of a carbon steel pipeline is 1.7×10 -7 ohm·m; Regarding the resistivity change caused by metal loss in the pitting area, a temperature compensation coefficient and a corrosion depth correction factor are introduced. For example, when the pitting depth is 2 mm, the local resistivity rises to 2.8×10 -7 ohm·m, and the correction coefficient is 1.65; The conductive characteristic data of the material is obtained through multi-point measurement, reflecting the influence of corrosion on the conductivity, providing a basis for subsequent analysis; In the corrosion area analysis, the area is divided into 5 cm×5 cm square grids, a total of 400 units are generated, and the equivalent resistance of each unit is calculated according to the corrosion depth. For example, the resistance increases by about 20% at a depth of 2 mm; The finite element method is used to solve the stray current distribution, and the boundary condition is set as the pipeline conductivity of 5.8×10 -7In Siemens per meter, iterative calculations show that the current density concentrates near the pitting pits, with a maximum value of 0.15 amperes per square centimeter, which is three times higher than that in the non-corroded area; By this method, the spatial distribution data of the current density is obtained, revealing the flow characteristics of the stray current in the corrosion area; Using a deep neural network to process the spatial distribution data of the current density, the network structure includes 2 convolutional layers and 3 fully connected layers. The input features are the grid cell coordinates and the equivalent resistance value of the grid cell, and the output is a continuous current density vector; The parameters are optimized by the backpropagation algorithm iteratively 8000 times to generate the current density distribution function, which can capture local abnormal changes better than the traditional interpolation method; At the same time, a bipolar potential gradient detector is used to scan along the corrosion boundary, with a probe spacing of 1 centimeter, and data is recorded every 5 centimeters. For example, the potential gradient in the severely pitted area reaches 2 volts per meter, while in the normal area it is only 0.3 volts per meter. A potential gradient spatial distribution map is generated by three-dimensional surface fitting, showing that the edge change rate exceeds 1 volt per meter per centimeter; The impedance change parameters are calculated according to the potential gradient spatial distribution map, including the impedance mean value, standard deviation, and maximum change rate. For example, the impedance mean value in the corrosion area is 35% higher than that in the normal area, the standard deviation is 0.08 ohms, and the change rate reaches 50% per meter; The impedance data spectrum is analyzed by 1024-point Fourier transform, and it is found that the high-frequency components are prominent, indicating that the impedance distribution is discontinuous; A mapping model between impedance characteristics and signal attenuation is established by using a support vector machine with a radial basis kernel function. Based on the training of 50 groups of corrosion samples, it is verified that when the local resistance increase index exceeds 1.5 and the impedance discontinuity index exceeds 2.0, the attenuation of the 1 kHz signal transmitted 100 meters reaches 45 dB, far exceeding the normal 25 dB, quantifying the impact of corrosion on electrical characteristics; In practical applications, if the pipeline materials are diverse, the grid size can be adjusted according to the material characteristics. For example, the stainless steel area can be reduced to 3 centimeters to improve the resolution; Dynamic scanning paths can be introduced in potential gradient detection to densify the measurement points along the high-value areas of the current density; The support vector machine model can also update the sample library regularly, incorporating new corrosion patterns to ensure that the evaluation accuracy is optimized with changes in operating conditions; Combined with the analysis of material resistivity and current density distribution, the law of corrosion distribution data is deduced. It is found that the resistance increases significantly around the pitting pits due to current concentration, which in turn causes impedance discontinuity and affects the stability of the grounding circuit; Through comprehensive evaluation, it is found that this change not only increases the local resistance, but also intensifies the reflection and attenuation of high-frequency signals. Targeted anti-corrosion measures need to be strengthened to ensure electrical performance; In practical applications, a 20% increase in resistance is the calculated conductivity value relative to the non-corroded area.

[0030] In one embodiment, the corrosion resistance increase criterion that the resistance increases by 20% means that the resistance value of the corroded area increases by 20% relative to the reference resistance value of the non-corroded area of the same material, and this criterion helps to analyze the law of corrosion distribution data.

[0031] In one embodiment, the law of corrosion distribution data is as follows: when the pitting depth > 1 mm, the local resistivity increase > 20%, the current density is concentrated at the corrosion edge, resulting in an impedance discontinuity index > 2.0. The local resistivity increase > 20% therein is relative to the corrosion resistance increase criterion.

[0032] As Figure 1 shown, in step S107, a pulse signal is sent to the traction network and the signal system, and the fault distance is calculated by analyzing the propagation time and amplitude attenuation data of the reflected waveform. At the same time, according to the evaluation result of the influence of corrosion and pitting on the electrical characteristics of the grounding path, the waveform interference offset is corrected to calibrate the fault point position.

[0033] Specifically, in step S107, a pulse signal generator is used to send a microsecond-level narrow pulse test signal to the traction network and the signal line. For example, the pulse width is set to 1 microsecond and the amplitude is 5 volts, and the reflected waveform is captured by a high-precision oscilloscope; To remove the influence of environmental noise, a band-pass filter is used to set the passband range from 10 Hz to 10 kHz to filter out high-frequency interference and generate initial reflected waveform data; In the noise reduction process, the adaptive threshold method is applied to the initial reflected waveform data, and the threshold is dynamically adjusted according to the real-time signal-to-noise ratio. For example, when the signal-to-noise ratio is lower than 10 dB, the threshold is set to 1.5 times the signal mean value; Combined with a median filter, the window length is set to 5 sampling points according to the noise characteristics to smooth the pulse interference. For example, the noise peak with an amplitude mutation exceeding 2 volts is eliminated to generate the reflected waveform after noise reduction; This method effectively retains the main features of the waveform while reducing the influence of interference on the positioning accuracy; Based on the corrosion depth and pitting distribution data, a transmission medium loss model is constructed. The attenuation coefficient is defined as 0.02 dB per meter, the dispersion coefficient is 0.01 microsecond per meter, and the reflection coefficient is 0.3; a three-layer neural network is used to calculate the waveform attenuation compensation value. The input layer includes the corrosion depth and the fault point distance parameter, where the distance parameter refers to the straight-line distance from the corrosion position to the measurement starting point, the hidden layer has 10 nodes, and the output layer is the corrected propagation delay; For example, when the corrosion depth of a certain path is 2 mm, the propagation delay increases by 0.05 microseconds, and after compensation, it is corrected to the true value, improving the distance calculation accuracy; Perform wavelet decomposition on the reflected waveform after noise reduction, select the db4 wavelet basis function to perform 4-layer decomposition, and extract the interference offset characteristics; Identify waveform distortion components through multi-scale analysis. For example, the high-frequency components in the first layer reveal noise residues, and the low-frequency components in the fourth layer reflect changes in the propagation path. Construct a time-frequency feature vector that includes time-domain and frequency-domain features; Subsequently, adopt an adaptive compensation algorithm. Calculate the phase offset angle and amplitude attenuation correction parameters according to the time-frequency feature vector. For example, the phase correction is 0.1 radian and the amplitude gain is 1.2 times. Generate a compensated reflected waveform to ensure that the waveform accurately reflects the characteristics of the fault point; In distance calculation, use the cross-correlation algorithm to analyze the time offset between the compensated reflected waveform and the transmitted signal. For example, when the peak correlation coefficient is 0.9, the propagation delay is 50 microseconds; Combined with the medium transmission speed, such as 3×10^8 m / s in the steel rail, calculate the initial fault distance to be 15 meters; For further calibration, construct a positioning error compensation model based on support vector regression. The input features include waveform attenuation rate, distortion degree, and propagation delay, and the output is the position calibration amount; The model is trained with 500 groups of samples and iteratively optimized until the mean square error is lower than 0.01 meter. For example, a certain calibration amount is 0.3 meter, and finally determine the fault point coordinates to be 15.3 meters; In practical applications, if the corrosion area is complex, the number of wavelet decomposition layers can be increased to 6 layers to capture more subtle distortions; The support vector regression model can introduce environmental temperature and humidity features to further optimize error compensation. Through this method, the positioning accuracy can be controlled within 1 meter, significantly improving the fault detection efficiency in complex environments. Combining multi-level signal processing and machine learning, not only accurately calculates the fault distance, but also effectively corrects the interference offset through corrosion impact assessment. In actual operation, this method can dynamically adjust parameters according to the line material and environment to ensure the reliability and practicality of the positioning results.

[0034] In one embodiment, collect the potential change data between the rail and the grounding grid in the subway traction DC power supply system, as well as the frequency characteristic data of the voltage signal in the AC power supply system. Determine whether the fault point is within the physical position interval output in step S104, send a pulse signal to the traction network and the signal line, and extract the propagation time and amplitude attenuation data from the reflected waveform reflected from the fault point to calculate the fault distance. The potential change data refers to the actual potential fluctuation data collected between the rail foundation and the grounding grid node. The frequency characteristic data refers to the data characterizing the spectral characteristics of the voltage signal in the AC power supply system. The reflected waveform is formed when a pulse signal is sent to the subway traction network. Due to impedance discontinuities (such as fault points) in the path, part of the pulse signal will be reflected back at these points. Analyzing the reflected waveform can help locate the fault point.

[0035] In one embodiment, the pitting corrosion distribution data source is generated by fusing pitting corrosion distribution data through ultrasonic grid scanning and infrared thermal imaging.

[0036] In one embodiment, neural network waveform compensation: a 3-layer fully connected network (input layer with 3 nodes: corrosion depth, distance, temperature; hidden layer with 10 nodes; output layer with 1 node: time delay compensation value), the optimizer is Adam, and the loss function is mean square error.

[0037] The present invention provides an integrated AC-DC grounding fault location system, mainly including: A potential acquisition module, used to acquire the potential change in the subway traction DC power supply system and the frequency characteristics in the AC power supply system, and determine the electrical signal characteristics of stray current corrosion; A signal decomposition module, used to perform 4-layer wavelet decomposition on the electrical signal using the db4 wavelet basis function in a strong electromagnetic interference environment, and extract the AC component characteristic values in the high-frequency component and the DC component characteristic values in the low-frequency component; A fault determination module, used to calculate the amplitude of the DC component characteristic value, and if it exceeds the preset DC amplitude threshold, determine a DC grounding fault. Calculate the amplitude, waveform distortion degree, and phase shift angle of the AC component characteristic value. If the amplitude exceeds the preset AC amplitude threshold and the waveform distortion degree > 3% or the phase shift angle > 30°, determine an AC grounding fault; A fault location module, used to combine the subway line topology structure, iteratively calculate the grid node potential values based on the finite element method, generate a potential gradient distribution map, calculate the current density between nodes through the current continuity equation, and determine the stray current flow vector in combination with the potential gradient direction; A corrosion analysis module, used to determine the physical position interval of the fault point based on the potential gradient distribution map and the stray current flow vector. If it is determined that there is a physical position interval of the fault point, then execute: measure the resistance between the pipeline or equipment shell and the preset grounding point, extract the corrosion distribution data in combination with the potential gradient distribution and the stray current flow vector, obtain the material resistivity, analyze the current density distribution and potential gradient change in the corrosion area, and evaluate the degree of local resistance increase, impedance discontinuity, and signal attenuation; A fault calibration module, used to send a narrow pulse signal with a pulse width ≤ 1 μs to the traction network and the signal system, calculate the initial fault distance according to the propagation time delay and amplitude attenuation data of the reflected waveform, and correct the waveform interference offset based on the evaluation results of the degree of local resistance increase, impedance discontinuity, and signal attenuation, and output the calibrated fault point coordinates.

[0038] For those skilled in the art, various corresponding changes and deformations can be made according to the above-described technical solutions and concepts, and all such changes and deformations should fall within the protection scope of the claims of this application.

Claims

1. An integrated AC-DC grounding fault location method, characterized in that, Including the following steps: Collect the potential change data of the subway traction DC power supply system and the frequency characteristic data of the AC power supply system, and determine the electrical signal characteristics of stray current corrosion; Under the strong electromagnetic interference environment, perform 4-layer wavelet decomposition on the electrical signal using the db4 wavelet basis function, and extract the AC component eigenvalue in the high-frequency component and the DC component eigenvalue in the low-frequency component; Calculate the amplitude of the DC component eigenvalue. If it exceeds the preset DC amplitude threshold, determine the DC grounding fault. Calculate the amplitude, waveform distortion degree, and phase shift angle of the AC component eigenvalue. If the amplitude exceeds the preset AC amplitude threshold and the waveform distortion degree > 3% or the phase shift angle > 30°, determine the AC grounding fault; Combined with the subway line topology structure, iteratively calculate the grid node potential value based on the finite element method, generate the potential gradient distribution map, calculate the current density between nodes through the current continuity equation, and determine the stray current flow vector in combination with the potential gradient direction; Based on the potential gradient distribution map and the stray current flow vector, determine the physical location interval of the fault point. If it is determined that there is a physical location interval of the fault point, execute: measure the resistance between the pipeline or equipment shell and the preset grounding point, extract the corrosion distribution data in combination with the potential gradient distribution and the stray current flow vector, obtain the material resistivity, analyze the current density distribution and potential gradient change in the corrosion area, and evaluate the degree of local resistance increase, impedance discontinuity, and signal attenuation; Send a narrow pulse signal with a pulse width ≤ 1 μs to the traction network and the signal system, calculate the initial fault distance according to the propagation delay and amplitude attenuation data of the reflected waveform, and correct the waveform interference offset based on the evaluation results of the degree of local resistance increase, impedance discontinuity, and signal attenuation, and output the calibrated fault point coordinates.

2. The AC-DC integrated grounding fault location method according to claim 1, wherein The extraction of the AC component eigenvalue in the high-frequency component and the DC component eigenvalue in the low-frequency component includes: The frequency band range of the high-frequency component is 250 Hz - 5 kHz, which is used to extract the AC component eigenvalue; The frequency band range of the low-frequency component is 0 - 250 Hz, which is used to extract the DC component eigenvalue.

3. The AC / DC integrated grounding fault location method according to claim 1, wherein The waveform distortion degree is extracted by calculating the fundamental wave and harmonic components through fast Fourier transform, and the phase shift angle is obtained by comparing the reference waveform through the cross-correlation algorithm.

4. The AC / DC integrated grounding fault location method according to claim 1, characterized in that, The convergence threshold for the iterative calculation of the grid node potential value is 0.001 V, the maximum number of iterations is 1000 times, and the spatial clustering of the stray current flow vector uses the K-means algorithm to identify the current inflow and outflow areas.

5. The AC / DC integrated grounding fault location method according to claim 1, characterized in that, The waveform interference offset correction includes: performing 4-layer decomposition of the db4 wavelet on the reflected waveform to extract the distortion component, and correcting the phase and amplitude through the adaptive compensation algorithm.

6. An integrated AC-DC grounding fault location system, characterized in that A potential acquisition module, which is used to collect the potential change in the subway traction DC power supply system and the frequency characteristics in the AC power supply system, and determine the electrical signal characteristics of stray current corrosion; A signal decomposition module, which is used to perform 4-layer wavelet decomposition on the electrical signal using the db4 wavelet basis function under the strong electromagnetic interference environment, and extract the AC component eigenvalue in the high-frequency component and the DC component eigenvalue in the low-frequency component; The fault determination module is used to calculate the amplitude of the DC component eigenvalue. If it exceeds the preset DC amplitude threshold, a DC grounding fault is determined. It calculates the amplitude, waveform distortion degree, and phase shift angle of the AC component eigenvalue. If the amplitude exceeds the preset AC amplitude threshold and the waveform distortion degree > 3% or the phase shift angle > 30°, an AC grounding fault is determined; The fault location module is used to combine the subway line topology structure, iteratively calculate the grid node potential value based on the finite element method, generate a potential gradient distribution map, calculate the current density between nodes through the current continuity equation, and determine the stray current flow vector in combination with the potential gradient direction; The corrosion analysis module is used to determine the physical location interval of the fault point based on the potential gradient distribution map and the stray current flow vector. If it is determined that there is a physical location interval of the fault point, the following operations are performed: measure the resistance between the pipeline or equipment shell and the preset grounding point, extract the corrosion distribution data in combination with the potential gradient distribution and the stray current flow vector, obtain the material resistivity, analyze the current density distribution and potential gradient change in the corrosion area, and evaluate the degree of local resistance increase, impedance discontinuity, and signal attenuation; The fault calibration module is used to send a narrow pulse signal with a pulse width ≤ 1 μs to the traction network and the signal system, calculate the initial fault distance according to the propagation delay and amplitude attenuation data of the reflected waveform, and correct the waveform interference offset based on the evaluation result of the degree of local resistance increase, impedance discontinuity, and signal attenuation, and output the calibrated fault point coordinates.

Images