Power grid fault detection and positioning method and system for smart grid system

Abstract

The application provides a power grid fault detection and positioning method and system for a smart grid system, and the method comprises the following steps: collecting a fault traveling wave signal when a power distribution line fails; preprocessing the fault traveling wave signal to obtain a preprocessed target fault traveling wave signal; detecting a traveling wave front of the target fault traveling wave signal to determine a signal arrival time of the fault traveling wave signal; extracting a fault signal feature from the target fault traveling wave signal through a frequency domain analysis method and a time domain analysis method; combining the fault signal feature and the signal arrival time to estimate a fault line characteristic of a fault line where the fault occurs; outputting a line type and a fault type of the fault line through a feature recognition model; selecting an optimal fault positioning algorithm based on the line type and the fault type and using a decision tree algorithm; and determining a line position of the fault line according to the target fault traveling wave signal and using the optimal fault positioning algorithm. The application has the effect of intelligently selecting an optimal positioning algorithm according to a line type.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of power grid fault detection, and particularly relates to a power grid fault detection and positioning method and system for a smart grid system. BACKGROUND

[0002] With the acceleration of urbanization and the growth of power demand, the structure of power grid becomes more and more complex. The line types in different regions and different application scenarios show a diversification trend, including high-voltage transmission lines, medium and low-voltage distribution lines, underground cables, and mixed lines. This complexity brings great challenges to fault location. At the same time, power users have continuously improved requirements for power supply reliability, requiring faster and more accurate fault location capabilities.

[0003] Since each type of line has its unique electrical characteristics and fault characteristics. Secondly, the same type of line will also show different characteristics under different environmental and operating conditions, which further increases the difficulty of selecting a suitable algorithm. The traditional fault location method is often designed for a specific type of line, lacking sufficient flexibility and adaptability, and is difficult to adapt to the characteristics of different line types, resulting in low positioning accuracy and low efficiency. SUMMARY

[0004] The application provides a power grid fault detection and positioning method and system for a smart grid system to solve the problem that the traditional fault location method is difficult to adapt to the characteristics of different line types, resulting in low positioning accuracy and low efficiency.

[0005] In a first aspect, the application provides a power grid fault detection and positioning method for a smart grid system, which comprises the following steps:

[0006] A traveling wave recording device deployed in the smart grid system is used to collect a fault traveling wave signal when a fault occurs on a distribution line;

[0007] The fault traveling wave signal is preprocessed to obtain a preprocessed target fault traveling wave signal;

[0008] The target fault traveling wave signal is subjected to traveling wave front edge detection to determine the signal arrival time of the fault traveling wave signal;

[0009] Fault signal characteristics are extracted from the target fault traveling wave signal by a frequency domain analysis method and a time domain analysis method;

[0010] The fault line characteristics of the fault line where the fault occurs are estimated in combination with the fault signal characteristics and the signal arrival time;

[0011] inputting the fault signal feature and the fault line characteristic into a pre-trained feature recognition model, outputting a line type and a fault type of the fault line through the feature recognition model;

[0012] selecting an optimal fault locating algorithm based on the line type and the fault type and using a decision tree algorithm;

[0013] determining a line position of the fault line according to the target fault traveling wave signal and using the optimal fault locating algorithm.

[0014] Optionally, the pre-processing the fault traveling wave signal to obtain a pre-processed target fault traveling wave signal comprises the following steps:

[0015] using a wavelet transform method to perform noise reduction processing on the fault traveling wave signal;

[0016] using a band-pass filter to perform filtering processing on the fault traveling wave signal after noise reduction processing to obtain a target fault traveling wave signal.

[0017] Optionally, the performing traveling wave front detection on the target fault traveling wave signal to determine a signal arrival time of the fault traveling wave signal comprises the following steps:

[0018] using a sliding window method to segment the target fault traveling wave signal into a plurality of traveling wave signal segments;

[0019] for each of the traveling wave signal segments, calculating to obtain a signal statistical feature and a signal first derivative of the traveling wave signal segment;

[0020] if the signal statistical feature does not exceed a pre-set feature threshold and / or the signal first derivative does not exceed a pre-set derivative threshold, it is determined that there is no traveling wave front in the traveling wave signal segment;

[0021] if the signal statistical feature exceeds the feature threshold and the signal first derivative exceeds the derivative threshold, it is determined that the traveling wave front exists in the traveling wave signal segment, and the traveling wave signal segment in which the traveling wave front exists is marked as a target traveling wave signal segment;

[0022] calculating to obtain the signal arrival time of the fault traveling wave signal based on the target traveling wave signal segment.

[0023] Optionally, the calculating to obtain the signal arrival time of the fault traveling wave signal based on the target traveling wave signal segment comprises the following steps:

[0024] improving a time resolution of the target traveling wave signal segment through an interpolation method;

[0025] According to a correlation analysis method, a signal cross-correlation function between the target traveling wave signal segment and a preset template signal is calculated;

[0026] A time corresponding to a function peak position of the signal cross-correlation function is taken as a signal arrival time of the fault traveling wave signal.

[0027] Optionally, the extracting of the fault signal feature from the target fault traveling wave signal by the frequency domain analysis method and the time domain analysis method comprises the following steps:

[0028] The target fault traveling wave signal is subjected to time domain analysis, and time domain features of the target fault traveling wave signal are counted, the time domain features comprising signal mean value, signal variance, signal skewness and signal kurtosis;

[0029] The target fault traveling wave signal is subjected to frequency domain analysis by Fourier transform, and frequency domain features of the target fault traveling wave signal are extracted, the frequency domain features comprising spectral distribution, harmonic content and power spectral density;

[0030] The time domain features and the frequency domain features are integrated into the fault signal feature of the target fault traveling wave signal.

[0031] Optionally, the feature recognition model comprises a line type recognition module and a fault type recognition module, the line type recognition module being constructed based on a support vector machine, and the fault type recognition module being constructed based on a convolutional neural network.

[0032] Optionally, the selecting of the optimal fault location algorithm based on the line type and the fault type and by using a decision tree algorithm comprises the following steps:

[0033] All basic line types and all basic fault types are converted into numerical features by one-hot encoding;

[0034] Information gain between all the numerical features is calculated;

[0035] All fault location algorithms are taken as leaf nodes, the information gain is taken as a splitting criterion, a CART algorithm is used to select a best numerical feature for splitting from a root node recursively until a number of node samples is less than a preset threshold, and an algorithm selection decision tree model is constructed;

[0036] The line type and the fault type are input into the algorithm selection decision tree model, and the optimal fault location algorithm is output by the algorithm selection decision tree model.

[0037] Optionally, the fault location algorithm comprises a traveling wave method, an impedance method, an injection method, a traveling wave-traveling wave correlation method and a multi-terminal synchronous measurement method.

[0038] In a second aspect, the present application also provides a power grid fault detection and positioning system for a smart grid system, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the power grid fault detection and positioning method for a smart grid system as described in the first aspect.

[0039] In a third aspect, the present application also provides a computer storage medium having a computer program stored thereon, wherein the computer program is executable by a processor to implement the steps of the power grid fault detection and positioning method for a smart grid system as described in the first aspect.

[0040] The present application has the following beneficial effects:

[0041] The present application can automatically select the optimal fault positioning algorithm by intelligently analyzing fault signal characteristics and line characteristics, greatly improving the accuracy and adaptability of fault positioning. This flexible method can effectively deal with various complex line structures, including overhead lines, underground cables, and hybrid lines, overcoming the limitations of traditional fixed algorithms when facing diversified lines. Secondly, the present application uses advanced signal processing techniques and artificial intelligence methods to extract more valuable information from fault traveling wave signals, improving the accuracy and efficiency of fault feature recognition. Thirdly, the present application introduces a decision tree algorithm to select the optimal fault positioning algorithm, which not only improves positioning efficiency but also continuously optimizes with the introduction of new lines and changes in grid structure, with good scalability and adaptability. In addition, the overall process design of the present application fully considers the actual application requirements, from signal acquisition, preprocessing to final positioning, forming a complete intelligent fault positioning solution, greatly improving the operation reliability and maintenance efficiency of the power system. In summary, the present application provides an innovative solution for fault positioning in complex grid environments, effectively solving the algorithm selection difficulty problem caused by complex line structures, and providing a strong guarantee for the safe and stable operation of the grid. BRIEF DESCRIPTION OF DRAWINGS

[0042] Figure 1 FIG. 1 is a flowchart of the power grid fault detection and positioning method for a smart grid system according to one embodiment of the present application. DETAILED DESCRIPTION

[0043] The technical solutions in the embodiments of the present application will be clearly described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by a person of ordinary skill in the art belong to the scope of protection of the present application.

[0044] The terms "first", "second", etc. in the specification and claims of the present application are used to distinguish similar objects, and are not used to describe a specific order or sequence. It should be understood that the data used in this way can be exchanged under appropriate circumstances, so that the embodiments of the present application can be implemented in an order other than those illustrated or described herein, and the objects distinguished by "first", "second", etc. are generally a class, and do not limit the number of objects, for example, the first object can be one or more. In addition, "and / or" in the specification and claims indicates at least one of the connected objects, and the character " / " generally indicates that the objects before and after are in an "or" relationship.

[0045] Figure 1 A flowchart of a power grid fault detection and positioning method for a smart grid system in an embodiment. It should be understood that, although Figure 1 The steps in the flowchart are displayed in sequence according to the direction of the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified herein, the execution of these steps is not strictly limited in sequence, and these steps can be executed in other orders. Moreover, Figure 1 At least part of the steps in the flowchart can include multiple sub-steps or multiple stages, which are not necessarily executed at the same time, but can be executed at different times, and the execution order of these sub-steps or stages is not necessarily sequential, but can be executed in rotation or alternation with at least part of other steps or sub-steps or stages of other steps. As Figure 1 The power grid fault detection and positioning method for a smart grid system disclosed by the present application specifically includes the following steps:

[0046] S101. Collecting a fault traveling wave signal when a fault occurs in a power distribution line by a traveling wave recording device deployed in a smart grid system.

[0047] Among the smart grid system, the traveling wave recording device is deployed, which is usually installed at the key nodes of the distribution line, such as substations, branch points, etc. When the distribution line fails, the fault point will produce high-frequency electromagnetic waves, which propagate along the line at the speed of light. The traveling wave recording device can capture these high-frequency signals and convert them into digital signals for storage and analysis. The traveling wave recording device usually uses an analog-to-digital converter with a high sampling rate (such as 10MHz or higher) to ensure that it can accurately capture rapidly changing traveling wave signals. The recorded signals usually include the instantaneous values of voltage and current, and sometimes also include zero sequence components. In order to improve the quality of the signal, the traveling wave recording device is usually equipped with a filter to remove power frequency signals and other low frequency interference. In addition, in order to ensure time synchronization, these devices are usually equipped with GPS clocks to achieve microsecond-level time accuracy. In actual application, multiple traveling wave recording devices may be deployed on a distribution line to improve positioning accuracy through multi-point measurement. For example, on a 10-kilometer-long distribution line, one traveling wave recording device may be installed at the starting end, the middle point and the terminal. When a fault occurs, these devices will trigger recording at the same time to capture the traveling wave signals generated by the fault. The raw signals collected are usually stored in high-precision digital format, such as 16-bit or 24-bit sampling precision, to preserve the details of the signal. These raw data are then transmitted to a central processing system for further analysis. In this way, high-quality, high-time-resolution fault traveling wave signals can be obtained, providing a reliable data basis for subsequent fault location analysis.

[0048] S102. Preprocess the fault traveling wave signal to obtain a preprocessed target fault traveling wave signal.

[0049] The main purpose of preprocessing is to remove noise and interference in the signal, improve the signal-to-noise ratio, and highlight the fault characteristics. First, data calibration is needed to convert the raw data into actual physical quantities, such as voltage (V) or current (A). This usually involves applying a predetermined conversion coefficient. Next, a digital filter is applied to remove unwanted frequency components. Common filtering methods include Butterworth filter, Chebyshev filter or Finite Impulse Response (FIR) filter. For example, a high-pass filter can be used to remove 50Hz or 60Hz power frequency components, with a cutoff frequency of about 1kHz. At the same time, a low-pass filter can be used to remove high-frequency noise, with a cutoff frequency of about 1MHz. After filtering, signal normalization may be needed to eliminate amplitude differences between different measurement points.

[0050] To further improve signal quality, wavelet transform can be applied for signal denoising. Wavelet transform enables simultaneous analysis of signals in time and frequency domains, making it particularly suitable for handling non-stationary signals. For example, db4 wavelet can be used for 5-level decomposition, and then soft thresholding method can be applied to high-frequency coefficients for denoising. The denoised signal is reconstructed by inverse wavelet transform to obtain clear target fault traveling wave signals. In addition, baseline correction can also be performed to eliminate the DC offset of the signal. This can be achieved by subtracting the mean value of the signal or using polynomial fitting. For example, a cubic polynomial can be used to fit the baseline of the signal, and then the fitted baseline is subtracted from the original signal.

[0051] The preprocessed target fault traveling wave signals should have higher signal-to-noise ratio and clearer fault features, providing a good foundation for subsequent front edge detection and feature extraction. Through these preprocessing steps, the accuracy and reliability of fault location can be significantly improved. The effect of preprocessing can be evaluated by comparing the time-domain waveforms and spectra of the original signal and the processed signal, ensuring that the fault features are enhanced while the noise and interference are effectively suppressed.

[0052] S103. Traveling wave front edge detection is performed on the target fault traveling wave signal to determine the signal arrival time of the fault traveling wave signal.

[0053] Wherein, the accuracy of this step directly affects the accuracy of subsequent fault location. Traveling wave front edge detection usually adopts a combination of multiple methods to improve the reliability of detection. First, the traditional threshold method can be used. This method sets a fixed or adaptive threshold, and when the signal amplitude exceeds this threshold, it is considered that the traveling wave front edge is detected. However, simple thresholding method may be affected by noise. Therefore, the difference method can be used in combination to improve the accuracy of detection. The difference method calculates the first or second difference of the signal, and the traveling wave front edge usually appears as a sudden change in the difference signal. Another effective method is to use a matched filter. Design a filter that matches the shape of the ideal traveling wave front edge, and then convolve it with the actual signal. The peak value of the convolution output corresponds to the position of the traveling wave front edge. The output of the matched filter can be represented as:

[0054]

[0055] Wherein, x[k] is the input signal, and h[n] is the impulse response of the matched filter.

[0056] Finally, to accurately determine the arrival time, interpolation techniques can be used. For example, cubic spline interpolation can be used near the detected front points to subdivide the sampling interval, obtaining sub-sampling precision arrival time estimates. Through the combined application of these methods, the traveling wave front can be accurately detected in various complex noise environments, and the arrival time of the signal can be determined. Accurate arrival time estimation is crucial for subsequent fault location, especially when using the time difference method for fault location. The typical detection accuracy can reach the microsecond level, which is crucial for accurately locating the fault point on the distribution line.

[0057] S104. Extracting fault signal features from the target fault traveling wave signal through frequency domain analysis method and time domain analysis method.

[0058] Among them, in terms of frequency domain analysis, Fast Fourier Transform (FFT) is mainly used to convert time domain signal to frequency domain. By observing the characteristics of the spectrum, the main frequency components of the fault signal can be identified. For example, single-phase ground fault usually produces obvious harmonic components near the fundamental frequency, while phase-to-phase short circuit fault may have characteristic peaks at higher frequencies. In terms of time domain analysis, the main features of the signal are amplitude, duration, rise time, etc. The mean, variance, skewness, kurtosis and other statistical characteristics of the signal can be calculated, which can reflect the overall distribution characteristics of the signal. At the same time, the envelope of the signal can also be extracted to analyze the attenuation characteristics of the signal. For transient faults, the duration and energy concentration of the signal can be calculated, which helps to distinguish different types of faults. In addition, Empirical Mode Decomposition (EMD) method can be used to decompose the signal into a series of Intrinsic Mode Functions (IMF), and then analyze the characteristics of each IMF. During the feature extraction process, attention should be paid to the selection of features that are distinguishable and stable to ensure that they can accurately reflect the characteristics of different types of faults. The extracted features usually include frequency domain features (such as the amplitude and phase of the main frequency components), time domain features (such as the peak value and root mean square value of the signal), and time-frequency joint features (such as energy distribution diagram). These features constitute a multi-dimensional feature vector, which provides an important basis for subsequent fault type recognition and location.

[0059] S105. Combining fault signal features and signal arrival time estimates to determine the fault line characteristics of the fault line.

[0060] Among them, first, according to the difference of signal arrival time, the approximate position of fault occurrence can be preliminarily judged. For example, if the signal arrival time difference of two adjacent measurement points is very small, then the fault point is likely to be located between the two measurement points. Second, using the frequency domain characteristics of the fault signal, the impedance characteristics of the fault point can be estimated. By analyzing the spectral relationship of the fault current and voltage, the equivalent impedance of the fault point can be calculated. This equivalent impedance information is very helpful for determining the fault type (such as low impedance ground fault or high impedance ground fault). Third, by analyzing the signal attenuation characteristics, the loss parameters of the line can be estimated. Specifically, the envelope line of the signal can be fitted using an exponential decay model, thereby obtaining the attenuation coefficient. This attenuation coefficient is related to the resistance, inductance and capacitance parameters of the line, and can be used to correct the line model. In addition, the signal propagation speed can also be used to estimate the distributed parameters of the line. The propagation speed of the traveling wave on the line is related to the unit length inductance and capacitance of the line, and the propagation speed can be calculated by the formula v=1 / √(LC), where L and C are the unit length inductance and capacitance of the line, respectively. By comparing the actual measured propagation speed with the theoretically calculated value, the line parameters can be corrected. Finally, combined with the polarity information and amplitude characteristics of the fault signal, the severity and duration of the fault can be estimated. These estimation results not only help fault location, but also provide reference for subsequent fault isolation and recovery strategies. It should be noted that the accuracy of the estimation results is affected by many factors, such as measurement error, nonlinear change of line parameters, etc., so in actual application, usually need to combine multiple methods for cross-validation to improve the reliability of the estimation.

[0061] S106. The fault signal characteristics and fault line characteristics are input into a pre-trained feature recognition model, and the line type and fault type of the fault line are output by the feature recognition model.

[0062] Among them, the feature recognition model includes a line type recognition module and a fault type recognition module. The line type recognition module is constructed based on a support vector machine (SVM), and the fault type recognition module is constructed based on a convolutional neural network (CNN). The SVM model can take the extracted feature vector as input, map the data to a high-dimensional space through a kernel function (such as a radial basis function), and then find the optimal classification hyperplane in the high-dimensional space. The CNN model is particularly suitable for processing time-frequency image features and can automatically learn local and global features of fault signals. In the model training phase, a large amount of known type fault data needs to be used for supervised learning. The training data should include samples of various line types (such as overhead lines, cables, hybrid lines, etc.) and fault types (such as single-phase grounding, two-phase short circuit, three-phase short circuit, etc.). In order to improve the generalization ability of the model, the cross-validation method is usually used to evaluate the model performance, and the regularization technique (such as L1, L2 regularization) is used to prevent overfitting. The output of the model is usually the probability distribution of the line type and the fault type, and the class with the highest probability can be selected as the final result. In order to improve the reliability of identification, the ensemble learning method such as Bagging or Boosting can be used to integrate the prediction results of multiple models. In addition, expert knowledge can be introduced to correct the output of the model by setting some rules, especially for some rare but important fault types. The performance of the model can be evaluated by accuracy, recall rate, F1 score, etc. In actual application, the model needs to be updated and fine-tuned regularly to adapt to the changes of the power grid system and the emergence of new fault types. Through accurate identification of line type and fault type, important basis can be provided for subsequent fault location and processing, and the efficiency and reliability of power grid operation and maintenance can be improved.

[0063] S107. Select the optimal fault location algorithm based on the line type and the fault type and using a decision tree algorithm.

[0064] Decision tree algorithms can automatically select the most suitable fault location algorithm based on different conditions. The construction of a decision tree is typically based on expert knowledge and historical data analysis. The root node of the tree can be the line type, and branches are formed based on factors such as fault type and line characteristics. For example, for a single-phase ground fault in an overhead line, a location algorithm based on the traveling wave method might be chosen; while for a phase-to-phase short-circuit fault in a cable line, the impedance method might be chosen. Each leaf node of the decision tree corresponds to a specific fault location algorithm. Commonly used fault location algorithms include the traveling wave method, the impedance method, and the injection method. The traveling wave method is suitable for rapidly changing transient faults, locating the fault point by calculating the time difference between the arrival of the fault traveling wave at different measurement points. Its mathematical model can be expressed as: x = (L + v(t2 - t1)) / 2, where x is the distance to the fault point, L is the total line length, v is the traveling wave propagation speed, and t1 and t2 are the times the traveling wave arrives at both ends. Impedance-based fault location methods are applicable to steady-state faults. They determine the fault location by calculating the impedance at the fault point, with the basic formula: x = (V / I)*(l / Z), where V and I are the fault point voltage and current, respectively, l is the line length, and Z is the impedance per unit length of the line. Injection methods locate faults by injecting a signal of a specific frequency into the line and analyzing its reflection characteristics; this method is particularly suitable for high-impedance faults. Decision tree algorithms consider not only fault type and line characteristics but also other factors such as the types of available measurement data, environmental conditions (e.g., weather), and historical fault statistics.

[0065] S108. Determine the location of the faulty line based on the target fault traveling wave signal and the optimal fault location algorithm.

[0066] The selected optimal algorithm is used to accurately calculate the location of the fault point.

[0067] The following are detailed implementation methods for several commonly used fault location algorithms:

[0068] 1. Traveling wave method:

[0069] This method locates faults based on the time difference of the traveling wave propagating along the line. Assuming there are two measurement points A and B, the total line length is L, the traveling wave speed is v, and the distance from the fault point to point A is x, then:

[0070]

[0071] Where tA and tB are the times when the fault traveling wave arrives at points A and B, respectively.

[0072] To improve accuracy, data from multiple measurement points can be used, and optimization can be performed using the least squares method:

[0073]

[0074] where xij is the fault distance calculated from measurement points i and j, \(\hat{x}_{ij}\) is the optimized estimated distance.

[0075] 2. Impedance method:

[0076] The impedance method uses fault point impedance to calculate fault distance. For single-phase ground fault, fault distance can be calculated by the following formula:

[0077]

[0078] where VF is the fault point voltage, IF is the fault current, Z1 is the positive sequence impedance, and L is the total length of the line.

[0079] To improve accuracy, the impact of load current can be considered, using an iterative algorithm:

[0080]

[0081] where IL is the load current, and xn is the fault distance estimate at the nth iteration.

[0082] 3. Injection method:

[0083] The injection method locates faults by injecting a signal of a specific frequency on the line and analyzing the reflected signal. Fault distance can be calculated by the following formula:

[0084]

[0085] where v is the signal propagation speed in the line, f is the frequency of the injected signal, and φ is the phase difference (degrees) between the incident and reflected waves.

[0086] 4. Traveling wave-traveling wave correlation method:

[0087] This method uses the correlation of traveling wave signals to improve positioning accuracy. First, calculate the cross-correlation function of the traveling wave signals measured at both ends:

[0088]

[0089] where sA(t) and sB(t) are the signals measured at A and B ends respectively. Fault distance can be calculated by the peak position τmax of the cross-correlation function:

[0090]

[0091] 5. Multi-terminal synchronous measurement method:

[0092] This method uses the synchronous data of multiple measurement points to improve positioning accuracy. Assuming there are n measurement points, the following equation set can be established:

[0093]

[0094] where (x, y) is the fault point coordinate, (xi, yi) is the coordinate of the ith measurement point, and ti is the time of the fault wave reaching the ith measurement point. This equation set can be solved by nonlinear least squares.

[0095] Through this comprehensive and adaptive fault location method, combined with advanced signal processing techniques and statistical analysis, the accuracy and reliability of fault location can be greatly improved. This not only helps to quickly find the fault point and reduce the outage time, but also provides accurate data support for preventive maintenance of the power grid, thereby improving the reliability and efficiency of the entire power distribution system.

[0096] In one embodiment, the pre-processing of the fault traveling wave signal to obtain the pre-processed target fault traveling wave signal includes the following steps:

[0097] Wavelet transform method is used to denoise the fault traveling wave signal;

[0098] A band-pass filter is used to filter the denoised fault traveling wave signal to obtain the target fault traveling wave signal.

[0099] In this embodiment, wavelet transform has good time-frequency localization characteristics and can analyze signals at different scales, making it particularly suitable for processing non-stationary signals such as fault traveling waves. In the implementation process, an appropriate wavelet basis function is first selected, commonly used ones include Daubechies wavelet and Symlet wavelet, and the selection needs to consider signal characteristics and computational complexity. Then the original signal is decomposed at multiple scales, usually to 3-5 layers. After decomposition, wavelet coefficients of different frequency bands are obtained, including low-frequency approximation coefficients and high-frequency detail coefficients. Next, threshold processing is performed on these coefficients to remove noise effects. The selection of threshold is crucial, and hard threshold method or soft threshold method can be used. The mathematical expression of hard threshold method is:

[0100]

[0101] where is the wavelet coefficient, is the threshold. The expression of soft threshold method is:

[0102]

[0103] The determination of threshold can use the universal threshold , where where N is the standard deviation of the noise, and N is the length of the signal. After processing, the signal is reconstructed using the inverse wavelet transform to obtain the denoised fault traveling wave signal. This method can effectively remove high-frequency noise while preserving the important characteristics of the signal, especially the sudden change characteristics at the fault instant. Compared with traditional filtering methods, wavelet denoising can better maintain the edge characteristics of the signal, reduce signal distortion, and improve the accuracy of subsequent analysis. In practical applications, the denoising effect can be optimized by adjusting the wavelet basis function, the number of decomposition layers, and the threshold selection strategy to adapt to different types of fault traveling wave signals.

[0104] A band-pass filter allows signals within a specific frequency range to pass through while suppressing frequency components outside that range, which is crucial for extracting the characteristic frequency band of the fault traveling wave. In designing a band-pass filter, the main frequency range of the fault traveling wave signal needs to be determined first, usually between a few kilohertz and a few hundred kilohertz. The type of filter can be selected from Butterworth, Chebyshev, or Elliptic filters, with Butterworth filters often being adopted due to their flat passband response. The order of the filter is usually selected between 4-8 to balance the filtering effect and computational complexity. The transfer function of the band-pass filter can be represented as:

[0105]

[0106] where K is the gain constant, and are the lower and upper cutoff frequencies, respectively. In digital implementation, a bilinear transformation can be used to convert the s-domain to the z-domain to obtain the difference equation of the digital filter. The filtering process can be implemented in the time domain through convolution or in the frequency domain through Fast Fourier Transform (FFT). The expression for time-domain implementation is:

[0107]

[0108] where and are the filter coefficients, and M and N are the filter orders. Frequency-domain implementation involves transforming the signal to the frequency domain using FFT, multiplying it by the filter frequency response, and then transforming it back to the time domain using inverse FFT. Band-pass filtering not only further removes residual noise but also highlights the characteristic frequency band of the fault traveling wave, enhancing the signal-to-noise ratio. This significantly improves the accuracy of subsequent fault feature extraction and positioning algorithms. In practical applications, the parameters of the band-pass filter can be dynamically adjusted according to different types of faults and line characteristics to achieve the best filtering effect.

[0109] In one embodiment, the target fault traveling wave signal is subjected to traveling wave front detection to determine the signal arrival time of the fault traveling wave signal, including the following steps:

[0110] The target fault traveling wave signal is segmented into multiple traveling wave signal segments using a sliding window method;

[0111] For each traveling wave signal segment, the signal statistical features and the signal first derivative of the traveling wave signal segment are calculated;

[0112] If the signal statistical features do not exceed the preset feature threshold and / or the signal first derivative does not exceed the preset derivative threshold, it is determined that there is no traveling wave front in the traveling wave signal segment;

[0113] If the signal statistical features exceed the preset feature threshold and the signal first derivative exceeds the derivative threshold, it is determined that there is a traveling wave front in the traveling wave signal segment, and the traveling wave signal segment with the traveling wave front is marked as a target traveling wave signal segment;

[0114] The signal arrival time of the fault traveling wave signal is calculated based on the target traveling wave signal segment.

[0115] In this embodiment, the target fault traveling wave signal is segmented into multiple traveling wave signal segments using a sliding window method. Specifically, a fixed-size window is slid on the time axis to divide the continuous signal into a series of overlapping short-time segments. The selection of the window size is usually based on the characteristic frequency and the expected duration of the fault traveling wave, and the typical value may be between a few milliseconds and a few tens of milliseconds. The sliding step determines the degree of overlap between adjacent windows. A smaller step can provide more detailed analysis, but will increase the amount of calculation. In actual operation, a window function such as a rectangular window, a Hanning window, or a Hamming window can be used to smooth the signal edges and reduce spectral leakage. For a signal x(n) of length N and a window function w(m) of length M, the kth signal segment can be represented as: where S is the sliding step, m=0,1,...,M-1. This segmentation method can capture the local feature changes of the signal, and is particularly suitable for analyzing non-stationary signals such as fault traveling waves, which helps to accurately locate the traveling wave front.

[0116] For each traveling wave signal segment, the signal statistical features and the signal first derivative of the traveling wave signal segment are calculated. The signal statistical features usually include mean, variance, skewness, and kurtosis, etc. The mean reflects the overall level of the signal, the variance describes the fluctuation degree of the signal, the skewness indicates the asymmetry of the distribution, and the kurtosis reflects the sharpness of the distribution. The signal first derivative reflects the rate of change of the signal, which can be calculated by difference approximation: These features and derivatives can effectively capture the mutation characteristics of the traveling wave front, providing a basis for subsequent judgment.

[0117] If the signal statistical feature does not exceed the preset feature threshold and / or the first-order derivative of the signal does not exceed the preset derivative threshold, it is determined that there is no traveling wave front in the traveling wave signal segment. This step determines whether there is a significant signal change by setting appropriate thresholds. The feature threshold is usually set based on the statistical characteristics of the signal in the normal operating state, and multiple standard deviations can be used as a reference. The derivative threshold reflects the upper limit of the signal change rate, which can be determined by analyzing historical fault data. If the calculated feature value and derivative do not exceed the corresponding threshold, it is considered that the signal segment is in a stable state and does not contain a traveling wave front. This method can effectively filter out normal fluctuations, reduce false positives, and improve the accuracy of traveling wave front detection. The selection of thresholds needs to balance sensitivity and false positive rate, which can be optimized and adjusted through experimental data.

[0118] If the signal statistical feature exceeds the feature threshold and the first-order derivative of the signal exceeds the derivative threshold, it is determined that there is a traveling wave front in the traveling wave signal segment, and the traveling wave signal segment with the traveling wave front is marked as the target traveling wave signal segment. This step combines multiple dimensions of judgment criteria to improve the reliability of detection. When the signal feature and the derivative exceed the threshold at the same time, it indicates that the signal not only has a significant change in amplitude, but also has a change rate that exceeds the normal range, which is a typical feature of the traveling wave front. The signal segment that meets the conditions is marked as the target traveling wave signal segment, and these segments contain key information of the fault traveling wave, which is crucial for subsequent fault location.

[0119] The signal arrival time is usually defined as the starting point of the traveling wave front and can be determined by various methods. One commonly used method is to find the first sampling point in the target traveling wave signal segment where the signal amplitude or energy exceeds a certain threshold. Another method is to use correlation analysis and perform cross-correlation operation on the target signal segment and the preset ideal traveling wave front template. The time when the correlation coefficient reaches the maximum value is the arrival time. After determining the arrival time, the accurate time stamp can be obtained in combination with the sampling rate. This time stamp is crucial for subsequent fault location calculations, especially when using the traveling wave method for fault location. Accurate arrival time can significantly improve the accuracy of fault location, reduce positioning errors, and speed up the fault recovery process.

[0120] In one embodiment, calculating the signal arrival time of the fault traveling wave signal based on the target traveling wave signal segment includes the following steps:

[0121] improve the time resolution of the target traveling wave signal segment by interpolation method;

[0122] calculate the signal cross-correlation function between the target traveling wave signal segment and the preset template signal according to the correlation analysis method;

[0123] the time corresponding to the function peak value of the signal cross-correlation function is taken as the signal arrival time of the fault traveling wave signal.

[0124] In this embodiment, the time resolution of the target traveling wave signal segment is improved by interpolation, which is an effective technique to enhance signal accuracy. This method inserts new data points between existing sampling points, thereby increasing the sampling rate and improving the time resolution. Common interpolation methods include linear interpolation, polynomial interpolation, and spline interpolation. Linear interpolation is the simplest method, assuming that the signal change between adjacent sampling points is linear. For a signal value y(t) between and , it can be calculated by the following formula:

[0125]

[0126] Higher-order interpolation methods such as cubic spline interpolation can provide smoother results, but the computational complexity will also increase. The interpolation multiple is usually selected between 2 and 10, depending on the original sampling rate and the required accuracy. This method can significantly improve the time resolution of the signal, helping to more accurately locate the traveling wave front, thereby improving the accuracy of fault location.

[0127] The cross-correlation function reflects the similarity between two signals as the time offset changes. For discrete signals x(n) and y(n), the cross-correlation function is defined as: where m is the time delay. In actual calculations, the Fast Fourier Transform (FFT) can be used to improve efficiency. The preset template signal is usually constructed based on the ideal traveling wave front characteristics and can be a step function, a ramp function, or a more complex waveform. Attention should be paid to the matching of signal length during calculation, and zero padding may be required for shorter signals. The calculation result of the cross-correlation function reflects the degree of matching between the target signal and the template signal as the time offset changes, providing a basis for accurately locating the traveling wave arrival time.

[0128] The peak of the cross-correlation function represents the position where the target signal and the template signal are most similar, which is likely to be the starting point of the traveling wave front. To improve accuracy, parabolic fitting or other interpolation methods can be used near the peak to obtain sub-sampling accuracy of the peak position. For example, using parabolic fitting can obtain a more accurate peak position through the following formula. After obtaining the peak position, combined with the sampling rate of the signal and the previous interpolation multiple, the accurate arrival time can be calculated. This method not only accurately locates the traveling wave front, but also to some extent, resists the influence of noise and interference, improving the reliability of fault location. Through this accurate time determination, the accuracy of subsequent fault location calculations can be significantly improved, providing important support for the rapid fault diagnosis and handling of power systems.

[0129] In one of the embodiments, the step of extracting the fault signal features from the target fault traveling wave signal by the frequency domain analysis method and the time domain analysis method comprises the following steps:

[0130] The time domain analysis is performed on the target fault traveling wave signal, and the time domain features of the target fault traveling wave signal are counted, the time domain features comprising a signal mean value, a signal variance, a signal skewness and a signal kurtosis;

[0131] The frequency domain analysis is performed on the target fault traveling wave signal by Fourier transform, and the frequency domain features of the target fault traveling wave signal are extracted, the frequency domain features comprising a frequency spectrum distribution, a harmonic content and a power spectral density;

[0132] The time domain features and the frequency domain features are integrated into the fault signal features of the target fault traveling wave signal.

[0133] In the embodiment, the time domain analysis is performed on the target fault traveling wave signal, and the time domain features of the target fault traveling wave signal are counted, which is a basic step of fault diagnosis. The time domain features directly reflect the change characteristics of the signal in the time dimension. The signal mean value μ reflects the overall level of the signal, and the calculation formula is , wherein N is the number of sampling points, and x_i is the value of the i th sampling point. The signal variance σ^2 describes the fluctuation degree of the signal around the mean value, and the calculation formula is . The signal skewness S reflects the asymmetry of the signal distribution, the positive skewness indicates that the right tail of the distribution is longer, and the negative skewness is opposite, and the calculation formula is . The signal kurtosis K describes the sharpness of the signal distribution, the kurtosis of the normal distribution is 3, greater than 3 indicates that the distribution is more sharp, and less than 3 indicates that the distribution is more flat, and the calculation formula is . These statistical features comprehensively reflect the overall shape and change characteristics of the fault traveling wave signal, which is helpful to distinguish different types of faults. In practical application, the time-varying values of these features can be calculated using the sliding window technology to capture the dynamic changes of the signal characteristics. The advantage of time domain analysis is intuitive and simple calculation, which can quickly provide the basic features of the fault signal, and provide an important basis for subsequent fault type judgment and positioning.

[0134] The frequency domain analysis is performed on the target fault traveling wave signal by Fourier transform, and the frequency domain features of the target fault traveling wave signal are extracted, which is a key step to deeply understand the nature of the fault signal. Fourier transform converts the time domain signal to the frequency domain, which reveals the frequency composition of the signal. For a discrete signal, the discrete Fourier transform (DFT) is used: where n = 0, 1,..., N-1, k = 0, 1,..., N-1. In practical computation, Fast Fourier Transform (FFT) algorithm is usually adopted to improve efficiency. The spectral distribution reflects the distribution of signal energy on each frequency component, which can be obtained by calculating the amplitude spectrum |X(k)|. Harmonic content analysis involves calculating the amplitude of integer multiple frequency components of the fundamental frequency, which can be used to evaluate the distortion degree of the signal. Power spectral density (PSD) describes how the signal power is distributed on the frequency, which can be estimated by periodogram method: During the analysis process, attention should be paid to selecting appropriate window functions (such as Hanning window) to reduce spectral leakage, and considering zero padding to improve frequency resolution. Frequency domain analysis can reveal periodic components and specific frequency characteristics in fault signals that may be ignored by time domain analysis, which is of great significance for identifying different types of faults (such as single-phase ground, inter-phase short circuit, etc.). In addition, frequency domain features are also very effective for detecting high-frequency transients and harmonic disturbances.

[0135] Integrating time domain features and frequency domain features into the fault signal features of the target fault traveling wave signal, this step fuses the information of time domain and frequency domain, forming a multi-dimensional feature vector, which provides comprehensive data support for subsequent fault diagnosis and classification. During the integration process, the scale and importance of different features need to be considered. Feature standardization methods such as z-score standardization can be used. This can eliminate the dimensional differences between different features. Another method is to use dimension reduction techniques such as Principal Component Analysis (PCA) or Independent Component Analysis (ICA) to extract the most representative feature combinations. For example, PCA can find the main variation direction of the data by solving the eigenvalues and eigenvectors of the covariance matrix. The integrated feature vector can be represented as where f_ti represents the time domain feature and f_fi represents the frequency domain feature. This comprehensive feature not only contains the statistical characteristics and frequency information of the signal, but also retains the potential correlation between time domain and frequency domain features. Through this integration, the characteristics of the fault signal can be more comprehensively described, improving the accuracy and reliability of fault diagnosis, and providing strong data support for subsequent fault type identification, positioning and severity assessment.

[0136] In one embodiment, selecting the optimal fault location algorithm based on line type and fault type and using decision tree algorithm includes the following steps:

[0137] Convert all basic line types and all basic fault types into numerical features using one-hot encoding;

[0138] Calculate the information gain between all numerical features;

[0139] All fault locating algorithms are taken as leaf nodes, and information gain is taken as the splitting criterion. The CART algorithm is used to recursively select the best numerical feature for splitting from the root node until the number of samples in the node is less than a pre-set threshold, and an algorithm selection decision tree model is constructed.

[0140] The line type and fault type are input into the algorithm selection decision tree model, and the optimal fault locating algorithm is output by the algorithm selection decision tree model.

[0141] In this embodiment, all basic line types and all basic fault types are converted into numerical features using one-hot encoding, which is a method of converting categorical variables into numerical inputs that can be directly used by machine learning algorithms. In this process, each categorical variable is converted into a binary vector whose length is equal to the number of categories, with only one element being 1 (representing that category) and the rest being 0. For example, if the line types include "overhead line", "cable", and "hybrid line", they can be encoded as [1, 0, 0], [0, 1, 0], and [0, 0, 1] respectively. Similarly, for fault types such as single-phase ground, two-phase short circuit, three-phase short circuit, etc., similar encoding can be used. The advantage of this encoding method is that it preserves the equidistance between categories and does not introduce any implicit order, so that each category is treated fairly. However, one-hot encoding can also lead to a sharp increase in the dimensionality of the feature space, especially when the number of categories is large. In practical applications, dimensionality reduction techniques or feature selection methods may need to be considered to handle high-dimensional features.

[0142] The calculation of information gain between all numerical features is a key step in the construction of a decision tree. Information gain is based on the concept of entropy in information theory and is used to measure the contribution of a feature to reducing the uncertainty of the system. For a feature A, its information gain can be expressed as IG(Y, A) = H(Y) - H(Y|A), where H(Y) is the entropy of the target variable Y, and H(Y|A) is the conditional entropy of Y given A. The formula for calculating entropy is H(X) = -Σp(x)log2p(x), where p(x) is the probability of event x. In actual calculations, for each feature, all possible values of the feature need to be traversed, and the reduction in system entropy when the feature takes each value is calculated, and then the average is taken as the information gain of the feature. For continuous features, discretization processing such as equal frequency binning or equal width binning is usually required. The process of calculating information gain can be time-consuming, especially when the number of features is large. Parallel computing techniques can be considered to speed up this process. The calculation result of information gain directly affects the structure of the decision tree, and a higher information gain means that the feature is more valuable for classification and should be selected as a splitting node.

[0143] All fault location algorithms are taken as leaf nodes, information gain is taken as the splitting criterion, and the CART algorithm is used to recursively select the best numerical feature for splitting from the root node until the number of node samples is less than a preset threshold. The construction of the algorithm selection decision tree model is an automatic process of selecting the optimal fault location algorithm. The CART algorithm is a binary tree algorithm that selects a feature and a threshold at each split to divide the dataset into two parts. At each node, the algorithm calculates the information gain of all possible splitting schemes and selects the feature and split point with the maximum information gain. This process is recursively performed until the stopping conditions are met, such as the number of node samples being less than a preset threshold or reaching the maximum tree depth. In this application, the leaf nodes represent different fault location algorithms (such as the traveling wave method, impedance method, etc.), rather than traditional class labels. The growth process of the tree is actually selecting the most suitable algorithm under different conditions. To prevent overfitting, pruning operations may be needed, such as using a validation set to evaluate the performance of each subtree and selecting the subtree with the strongest generalization ability.

[0144] The line type and fault type are input into the algorithm selection decision tree model, and the optimal fault location algorithm is output by the algorithm selection decision tree model, which is the last step of the entire fault location process and the key link of applying theoretical models to practical problems. In this step, first, the actual line type and fault type are converted into numerical features through the one-hot encoding method established earlier. Then, these features are input into the trained decision tree model. The decision tree model will start from the root node and pass down layer by layer according to the splitting rules of each internal node until it reaches a leaf node. This leaf node corresponds to the optimal fault location algorithm for the current input condition. For example, for a single-phase ground fault of an overhead line, the model may choose the traveling wave method as the optimal algorithm; for a two-phase short-circuit fault of a cable line, the model may choose the impedance method. This automated algorithm selection process not only makes quick decisions but also adjusts flexibly according to actual conditions, adapting to different fault scenarios. At the same time, due to the characteristics of the decision tree model, the decision path can be easily traced to understand why a particular algorithm was chosen, which is helpful for the interpretability of the results and the maintainability of the system.

[0145] The application further discloses a power grid fault detection and positioning system for a smart grid system, which comprises a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor implements the power grid fault detection and positioning method for the smart grid system as described in any one of the above embodiments when executing the computer program.

[0146] The processor can be a central processing unit (CPU), and can also be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), programmable logic devices (PLD), field programmable gate arrays (FPGA), or other programmable logic devices, discrete gate or transistor logic, discrete hardware components, etc.

[0147] The memory can be an internal storage unit of the computer device, for example, a hard disk or a memory of the computer device, or an external storage device of the computer device, for example, a plug-in hard disk, a smart memory card (SMC), a secure digital card (SD) or a flash memory card (FC) equipped on the computer device, or a combination of the internal storage unit and the external storage device of the computer device.

[0148] The application further discloses a computer readable storage medium, which stores instructions, and the instructions enable the processor to be configured to execute the power grid fault detection and positioning method for the smart grid system described in any one of the embodiments when executed by the processor.

[0149] The computer program can be stored in a machine readable medium, and the computer program includes computer program code, which can be in the form of source code, object code, executable files or some intermediate forms, etc., and the machine readable medium includes any entity or device, recording medium, U disk, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier wave signal, telecommunication signal and software distribution medium, etc. that can carry the computer program code, and it should be noted that the machine readable medium includes but is not limited to the above-mentioned components.

[0150] The computer readable storage medium stores the power transmission line comprehensive fault detection method in the computer readable storage medium, and is loaded and executed on the processor, so as to facilitate the storage and application of the method.

[0151] Those skilled in the art should understand that the above discussion of any of the embodiments is merely exemplary and is not intended to be limiting of the scope of protection as set forth in the appended claims; the above embodiments or technical features among different embodiments can also be combined, steps can be implemented in any order, and there are many other changes, such as the above-mentioned different aspects of one or more embodiments of the present application, which are not provided in detail for the sake of brevity.

[0152] One or more embodiments of the present application are intended to cover all such alternatives, modifications, and variations as fall within the broad scope of the application. Accordingly, any and all such alternatives, modifications, equivalents, improvements and the like as are deemed by those of ordinary skill in the art to fall within the spirit and scope of one or more embodiments of the present application are intended to be included within the scope of the present application.

Claims

1. A method for detecting and locating power grid faults in a smart grid system, characterized in that, The method includes the following steps: Fault traveling wave signals are collected when a power distribution line experiences a fault by using traveling wave recording devices deployed in the smart grid system. The fault traveling wave signal is preprocessed to obtain the preprocessed target fault traveling wave signal; The rising edge of the target fault traveling wave signal is detected to determine the signal arrival time of the fault traveling wave signal. Fault signal features are extracted from the target fault traveling wave signal using frequency domain analysis and time domain analysis methods. The characteristics of the faulty line are estimated by combining the fault signal characteristics and the signal arrival time. The fault signal features and the fault line characteristics are input into a pre-trained feature recognition model, and the line type and fault type of the fault line are output through the feature recognition model. Based on the line type and the fault type, the optimal fault location algorithm is selected using a decision tree algorithm; The location of the faulty line is determined based on the target fault traveling wave signal and the optimal fault location algorithm.

2. The method for detecting and locating power grid faults in a smart grid system according to claim 1, characterized in that, The preprocessing of the fault traveling wave signal to obtain the preprocessed target fault traveling wave signal includes the following steps: The fault traveling wave signal is denoised using wavelet transform. The noise-reduced fault traveling wave signal is filtered using a bandpass filter to obtain the target fault traveling wave signal.

3. The method for detecting and locating power grid faults in a smart grid system according to claim 1, characterized in that, The step of detecting the leading edge of the target fault traveling wave signal and determining the signal arrival time of the fault traveling wave signal includes the following steps: The target fault traveling wave signal is segmented into multiple traveling wave signal segments using the sliding window method; For each traveling wave signal segment, the signal statistical characteristics and the first derivative of the traveling wave signal segment are calculated. If the statistical features of the signal do not exceed the preset feature threshold and / or the first derivative of the signal does not exceed the preset derivative threshold, then it is determined that there is no traveling wave leading edge in the traveling wave signal segment; If the statistical features of the signal exceed the feature threshold and the first derivative of the signal exceeds the derivative threshold, then it is determined that the traveling wave leading edge exists in the traveling wave signal segment, and the traveling wave signal segment with the traveling wave leading edge is marked as the target traveling wave signal segment. The arrival time of the fault traveling wave signal is calculated based on the target traveling wave signal segment.

4. The method for detecting and locating power grid faults in a smart grid system according to claim 3, characterized in that, The step of calculating the arrival time of the fault traveling wave signal based on the target traveling wave signal segment includes the following steps: The temporal resolution of the target traveling wave signal segment is improved by interpolation. The cross-correlation function between the target traveling wave signal segment and the preset template signal is calculated using the correlation analysis method. The time corresponding to the peak position of the cross-correlation function of the signal is taken as the arrival time of the fault traveling wave signal.

5. The method for detecting and locating power grid faults in a smart grid system according to claim 1, characterized in that, The step of extracting fault signal features from the target fault traveling wave signal using frequency domain analysis and time domain analysis methods includes the following steps: The target fault traveling wave signal is subjected to time-domain analysis, and the time-domain characteristics of the target fault traveling wave signal are statistically analyzed. The time-domain characteristics include signal mean, signal variance, signal skewness, and signal kurtosis. The target fault traveling wave signal is analyzed in the frequency domain by Fourier transform, and the frequency domain features of the target fault traveling wave signal are extracted. The frequency domain features include spectral distribution, harmonic content and power spectral density. The time-domain features and the frequency-domain features are integrated into the fault signal features of the target fault traveling wave signal.

6. The method for detecting and locating power grid faults in a smart grid system according to claim 1, characterized in that, The feature recognition model includes a line type recognition module and a fault type recognition module. The line type recognition module is built based on a support vector machine, and the fault type recognition module is built based on a convolutional neural network.

7. The method for detecting and locating power grid faults in a smart grid system according to claim 1, characterized in that, The step of selecting the optimal fault location algorithm based on the line type and the fault type using a decision tree algorithm includes the following steps: One-hot coding is used to convert all basic line types and all basic fault types into numerical features; The information gain among all the numerical features is calculated; All fault location algorithms are treated as leaf nodes. Using the information gain as the splitting criterion, the CART algorithm is used to recursively select the best numerical features for splitting from the root node until the number of node samples is less than a preset threshold, thus constructing an algorithm selection decision tree model. The line type and the fault type are input into the algorithm selection decision tree model, and the optimal fault location algorithm is output through the algorithm selection decision tree model.

8. The method for detecting and locating power grid faults in a smart grid system according to claim 7, characterized in that, The fault location algorithms include traveling wave method, impedance method, injection method, traveling wave-traveling wave correlation method, and multi-terminal synchronous measurement method.

9. A power grid fault detection and location system for a smart grid system, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the power grid fault detection and location method for smart grid systems as described in any one of claims 1 to 8.

10. A computer storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the power grid fault detection and location method for a smart grid system as described in any one of claims 1 to 8.

Images