A ground cable high-frequency signal detection method, system, device and medium

Abstract

A grounding cable high-frequency signal detection method, system, device and medium adopt compressed sensing (CS) technology, through sparse representation and compressed sampling of cable reflection signal, the amount of sampling data is significantly reduced, and through the sparse signal reconstruction algorithm, the efficient reconstruction of the cable reflection signal is realized, the problems of slow signal processing speed, poor anti-interference ability and low detection precision in the prior art are solved; the system is composed of a signal acquisition module, a compressed sampling module, a signal reconstruction module and a fault detection and positioning module, and can realize accurate judgment of the type and position of the cable fault in a complex environment, and is especially suitable for online monitoring of long-distance cables; the device and medium detect the grounding cable high-frequency signal based on the above detection method and system, compared with the existing time domain reflection method (TDR), the application has higher detection sensitivity, anti-interference ability and processing speed, and can complete cable fault diagnosis and attenuation compensation without reference data, and provides a fast and reliable solution for cable state monitoring.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of cable state detection, and in particular to a grounding cable high-frequency signal detection method, system, device and medium. BACKGROUND

[0002] With the development of power systems, cables, as an important component of power transmission networks, are widely used in urban power supply, industrial and mining enterprises, and transportation infrastructure. Among them, grounding cables are important components to ensure the safe operation of equipment, and their health status is directly related to the stability and reliability of the power system. However, during long-term operation, cables may be affected by moisture, aging, mechanical damage and external environmental stress, which may cause insulation layer aging, partial discharge and other hidden dangers, and ultimately lead to cable failure. Therefore, it is of great significance to carry out state monitoring and fault detection of grounding cables.

[0003] Traditional cable detection methods such as time domain reflectometry (TDR) and frequency domain reflectometry (FDR) mainly detect fault locations in cables based on the propagation characteristics of low-frequency signals. However, due to the complex arrangement structure of grounding cables and external interference, low-frequency signals are prone to attenuation and reflection waveform overlap in long-distance transmission, resulting in a decrease in detection accuracy and sensitivity. In addition, as the length of the cable increases, the resolution capability of low-frequency signals is also limited, making it difficult to effectively identify multiple fault points or complex insulation defects. In contrast, high-frequency signals have shorter wavelengths and higher resolution, which can effectively avoid signal attenuation and multiple reflection problems in long cable transmission paths. Therefore, using high-frequency signals for grounding cable detection has become a potential technical means.

[0004] Current high-frequency signal detection methods require uniform and dense sampling of the entire frequency range. In long-distance cables or complex environments, the time and resource requirements for sampling and processing increase significantly, increasing the cost of the detection system and reducing detection efficiency. Direct sampling of high-frequency signals in the cable follows the Nyquist sampling theorem, which requires the sampling frequency to be at least twice the highest frequency of the signal. In long-distance cables, the required sampling time is longer, resulting in a large amount of sampling data, which in turn requires more storage space to store sampling data and more powerful processors to process. Rao Xianjie, Xu Zhonglin, Ding Yuqin, Hu Xiaoyu, Zhou Kai. Cable defect positioning method based on short-time fractional Fourier transform and time-frequency domain reflectometry [J / OL]. Electrical Engineering Technology Journal. A cable defect positioning method based on short-time fractional Fourier transform (STFRFT) and traditional time-frequency domain reflectometry (TFDR) is proposed to solve the problems of low time-frequency resolution and severe cross-term interference in traditional time-frequency domain reflectometry (TFDR). These methods still have the problem of excessive sampling data for different application objects. SUMMARY

[0005] In order to overcome the above-mentioned deficiencies of the prior art, the present application provides a grounding cable high-frequency signal detection method, system, device and medium, which adopts compressed sensing (CS) technology, realizes efficient detection and fault diagnosis of cable signals through sparse representation and compressed sampling of cable reflection signals. This method can greatly reduce the number of data acquisition, while improving the sensitivity and accuracy of detection.

[0006] In order to achieve the above-mentioned purpose, the specific technical scheme adopted by the present application is as follows:

[0007] A grounding cable high-frequency signal detection method, specifically comprising the following steps:

[0008] Step 1, injecting a detection signal into the cable, generating partial reflection signals at the discontinuous points in the cable, and capturing the reflected signals by a cable reflection signal acquisition module;

[0009] Step 2, performing sparse transformation on the reflected signals captured by the cable reflection signal acquisition module, extracting key features in the signals, including time difference, amplitude change, attenuation coefficient and frequency domain response of the reflection signals, which can reflect the cable state or fault position information, removing the redundant information in the signals, and providing a sparse part containing signal features, i.e. sparse signals, for the compressed sampling module;

[0010] Step 3, constructing a Gaussian measurement matrix, randomly projecting the sparse signals provided in step 2, realizing compressed sampling of the sparse signals, and outputting the compressed sparse signals;

[0011] Step 4, obtaining the reconstructed original signal containing cable fault and positioning information by sparse signal reconstruction algorithm on the compressed sparse signals in step 3

[0012] Step 5, performing spectrum analysis on the reconstructed original signal output in step 4 to obtain the characteristics of the reflection signals;

[0013] Step 6, according to the characteristics of the reflection signals obtained in step 5, including reflection amplitude, phase change and pulse width, identifying the fault signals, and judging the cable fault type, cable switch state and switch operation.

[0014] The specific method of step 2 is:

[0015] The reflected signal captured by the cable reflection signal acquisition module in step 1 is denoted as sparse signal The sparse signal ​Sparse representation is performed using the discrete wavelet transform DWT as θ = DWT(x), where θ is the sparse coefficient in the wavelet basis; the matrix expression of the discrete wavelet transform DWT is the wavelet basis matrix, which is composed of wavelet basis functions; the wavelet basis functions include the mother wavelet function ψ k , , , , k , ,

[0022] , (t) = 2 j / 2 ψ(2 j t - k) and the scaling function where j represents the scale and k represents the translation; for the discrete wavelet transform of the cable reflection signal x with length N, the wavelet basis matrix is:

[0016] is an N×N matrix, ψ0(t k ) represents the value of the mother wavelet function at different scales j and sampling points t k underneath, represents the value of the scaling function at different scales j and sampling points t k underneath;

[0017] Then, a threshold ε is set, and the coefficients less than ε are set to zero to obtain the sparsified coefficients The sparse signal x is represented as x = Aθ, where A is the wavelet basis matrix representing the cable reflection signal and θ is the sparse coefficient in the wavelet basis.

[0018] The specific method of step 3 is as follows:

[0019] Construct a random M×N - dimensional Gaussian measurement matrix Ф, and the wavelet basis matrix is where M << N, and each element φ ij in the random Gaussian measurement matrix Ф is a random variable, following a Gaussian distribution with a mean of 0 and a variance of 1: Through the random Gaussian measurement matrix Ф, the sparse signal x in step 2 is randomly projected to obtain the compressed signal y = Фx, reducing the number of sampling points. [[ID=​​​​​​​​​​​​​​​​​​​i It is the i-th column of the wavelet basis matrix A, representing the i-th wavelet basis vector;

[0023] 2) Index the selected wavelet basis vectors Add to index collection S k Chinese: S k+1 =S k ∪{i k};

[0024] 3) Use the least squares method in the current index set S k+1 The optimal sparse coefficients of the reconstructed signal are calculated above, i.e.

[0025]

[0026] in, The submatrix formed by the selected basis vectors;

[0027] 4) Update residuals

[0028] 5) Set the residual threshold ε1, and the upper limit of the number of iterations is K;

[0029] 6) When the residual r k+1 The algorithm stops when the number of iterations is less than the preset threshold ε1, or when the number of iterations reaches K.

[0030] 7) Output the reconstructed original signal containing cable fault and location information.

[0031] The specific method for step 5 is as follows:

[0032] The reconstructed original signal is obtained through step 4. Calculate the reflection coefficient at different injected signal frequencies:

[0033] 1) Reconstruct the original signal Transform to the frequency domain V(f) using Laplace transform;

[0034] 2) The transformed frequency domain V(f) includes the injected signal V(f). zhuru and the reflected signal V(f) fans That is, V(f) = V(f) fans +V(f) zhuru Then, the reflection coefficient at different injected signal frequencies is calculated. Among them, V fans (f) is the amplitude of the reflected signal, V zhuru (f) is the amplitude of the injected signal;

[0035] By injecting different signal frequencies, the characteristics of the reflected signal, namely the spectrum formed by the change of the reflection coefficient with frequency, are obtained.

[0036] The specific method of step 6 is:

[0037] According to step 5, the frequency response curve is established by measuring the change of the reflection coefficient under different signal frequencies; the reflection power coefficient R = |Γ| is obtained from the reflection coefficient 2 The fault type and switch state are judged by the reflection coefficient and the reflection power coefficient.

[0038] The fault type and switch state are judged by the reflection coefficient and the reflection power coefficient, and the specific method is as follows:

[0039]

[0040] A ground cable high-frequency signal detection system, comprising:

[0041] A cable reflection signal acquisition module is used to capture the partial reflection signal generated by the detection signal injected into the cable at the discontinuous point in the cable in step 1;

[0042] A sparse signal acquisition module is used to perform sparse transformation on the reflection signal captured in step 1 in step 2, extract key features in the signal, remove redundant information in the signal, and obtain a sparse part containing signal features;

[0043] A compressed sampling module is used to construct a Gaussian measurement matrix in step 3, and randomly project the sparse signal obtained in step 2 to realize compressed sampling of the sparse signal, and output the compressed sparse signal;

[0044] A signal reconstruction module is used to output a reconstructed original signal containing cable fault and positioning information by a sparse signal reconstruction algorithm in step 3 in step 4;

[0045] A fault detection and positioning module is used to perform spectral analysis on the reconstructed original signal output in step 4 in step 5 to obtain the characteristics of the reflection signal;

[0046] A fault type and switch state judgment module is used to identify the fault signal according to the characteristics of the reflection signal obtained in step 5 in step 6, and judge the cable fault type and cable switch state and switch operation.

[0047] A ground cable high-frequency signal detection device, comprising: a memory and a processor, the memory stores a computer program, so that the processor executes the ground cable high-frequency signal detection method.

[0048] A receiving user input program storage medium, the stored computer program is executed by the processor to realize efficient detection and fault diagnosis of the cable signal based on the ground cable high-frequency signal detection method.

[0049] Compared with the prior art, the present application has the following beneficial effects:

[0050] 1. The present application adopts the compressed sampling technology, performs sparse transformation on the captured reflection signal, extracts the key features in the signal, removes the redundant information in the signal, effectively reduces the amount of sampling data, reduces the storage and transmission pressure, and at the same time maintains the high-precision detection of the fault position and type. This feature makes the application of the technology in large-scale monitoring system more efficient.

[0051] 2. By introducing the compressed sensing (CS) and signal sparse representation method, the cable reflection signal is more efficiently collected and reconstructed, and the detection sensitivity and accuracy of the cable fault are improved, especially in long-distance cable and complex environment, which can effectively cope with signal attenuation and noise interference.

[0052] 3. In the signal processing process, the present application adopts an optimized reconstruction algorithm to obtain a reconstructed original signal containing cable fault and positioning information, thereby reducing the computational complexity.

[0053] 4. In the detection process, the present application does not need to rely on the reference data of the complete cable, and can directly detect the cable fault and attenuation compensation on site. Compared with the traditional FDR method, the limitation of needing intact cable reference test data is solved, and the practicability and flexibility are improved.

[0054] In summary, the detection method of the present application adopts the compressed sensing (CS) technology, realizes the efficient detection and fault diagnosis of the cable signal by sparse representation and compressed sampling of the cable reflection signal, and has obvious advantages in detection accuracy, processing speed and anti-interference ability, effectively solving the deficiencies in the prior art. BRIEF DESCRIPTION OF DRAWINGS

[0055] To more clearly illustrate the embodiments of the present application, the drawings required for the embodiments will be briefly introduced below, and the drawings are only some embodiments of the present application. Those skilled in the art can obtain other drawings according to these drawings without creating any creative labor.

[0056] Figure 1 A high-frequency signal detection method principle block diagram for the present application is used. DETAILED DESCRIPTION

[0057] The present application will be further described below in conjunction with the embodiments, but does not constitute any limitation on the present application, and any limited number of modifications made by anyone within the scope of the claims of the present application is still within the scope of the claims of the present application.

[0058] As shown in Figure 1 , the relationship between the voltage phasor on the grounding cable and the current phasor flowing through the grounding cable and the resistance and reactance of the grounding cable.

[0059] A grounding cable high-frequency signal detection method, specifically comprising the following steps:

[0060] Step 1, the detection signal injected into the cable generates a partial reflection signal at the discontinuous point in the cable, and the reflected signal is captured by the cable reflection signal acquisition module;

[0061] Step 2, sparse transform is performed on the reflected signal captured by the cable reflection signal acquisition module, and key features in the signal are extracted, including time difference, amplitude change, attenuation coefficient and frequency domain response of the reflected signal, which can reflect the information of the cable state or fault position, and the redundant information in the signal is removed, and the sparse signal containing the sparse part of the signal features is provided for the compressed sampling module; The specific steps are as follows:

[0062] The signal captured by the cable reflection signal acquisition module in step 1 is denoted as sparse signal Discrete wavelet transform DWT is applied to the sparse signal for sparse representation θ=DWT(x), wherein θ is the sparse coefficient under the wavelet basis; the matrix expression of discrete wavelet transform DWT is the wavelet basis matrix composed of wavelet basis functions; the wavelet basis function includes mother wavelet function j,k (t)=2 j / 2 ψ(2 j t-k) and scale function Where j represents the scale, and k represents the translation; for the discrete wavelet transform of the cable reflection signal x with length N, the wavelet basis matrix is:

[0063] is an N×N matrix, and k ψ0(t k ) represents the value of the mother wavelet function at different scales j and sampling points t k , and represents the value of the scale function at different scales j and sampling points t k .

[0064] Then, set the threshold value ε, and set the coefficients less than ε to zero to obtain the sparse coefficient The sparse signal x is expressed as x=Aθ, A is the wavelet basis matrix representing the cable reflection signal, and θ is the sparse coefficient under the wavelet basis.

[0065] Step 3, the compressed sampling module performs random projection on the sparse signal provided in step 2 through a pre-set measurement matrix, realizes the compressed sampling of the sparse signal, and the compressed sparse signal enters the signal reconstruction module; The specific is as follows:

[0066] A random M x N dimensional Gaussian measurement matrix Ф is constructed, and the wavelet basis matrix is where M << N, each element φ ij in the random Gaussian measurement matrix Ф is a random variable, which follows a Gaussian distribution with mean 0 and variance 1: The sparse signal x in step 2 is randomly projected by the random Gaussian measurement matrix Ф to obtain the compressed signal y = Фx, thereby reducing the number of sampling points.

[0067] Step 4, the signal reconstruction module obtains the reconstructed original signal containing cable fault and positioning information by a sparse signal reconstruction algorithm such as orthogonal matching pursuit (OMP) and the like on the compressed sparse signal in step 3 The reconstructed original signal is transmitted to the fault detection and positioning module; the details are as follows:

[0068] According to the compressed signal y obtained in step 3 and the wavelet basis matrix The compressed signal y is set to be the original signal sampled by the Gaussian measurement matrix Ф, the residual signal r0 is initialized as the compressed signal y, r0 = y, and the residual is the original observation signal; the index of the wavelet basis vector is selected The most relevant basis vector is selected from the wavelet basis matrix A and the current residual r k , as follows:

[0069] 1) Select the basis vector with the highest correlation: find the index i that maximizes , that is, k where A i is the i-th column of the wavelet basis matrix A, representing the i-th wavelet basis vector;

[0070] 2) Add the index i of the selected wavelet basis vector to the index set S k : S k+1 = S k ∪{i k};

[0071] 3) Calculate the optimal sparse coefficient of the reconstructed signal on the current index set S k+1 using the least squares method, that is,

[0072]

[0073] where A is a submatrix composed of the selected basis vectors;

[0074] 4) Update the residual

[0075] 5) Set the residual threshold ε1 and the upper limit of the iteration number K;

[0076] 6) When the residual r k+1 is less than a preset threshold ε1, or the number of iterations reaches K, the algorithm stops;

[0077] 7) Output the reconstructed original signal containing cable fault and positioning information

[0078] Step 5, the fault detection and positioning module performs spectral analysis on the reconstructed original signal output in step 4 to obtain the characteristics of the reflected signal and identify possible fault signals; the specific steps are as follows:

[0079] The reconstructed original signal is obtained by step 4 The reflection coefficient under different injected signal frequencies is calculated:

[0080] 1) Reconstruct the original signal Convert to the frequency domain V(f) by Laplace transform;

[0081] 2) The converted frequency domain V(f) contains the injected signal V(f) zhuru and the reflected signal V(f) fans , that is, V(f) = V(f) fans + V(f) zhuru ; Then calculate the reflection coefficient under different injected signal frequencies Where V fans (f) is the reflected signal amplitude, and V zhuru (f) is the injected signal amplitude;

[0082] By injecting different signal frequencies, the characteristics of the reflected signal, that is, the graph formed by the change of the reflection coefficient with frequency, are obtained.

[0083] Step 6, the fault type and switch state judgment module judges the cable fault type and cable switch state and switch operation according to the characteristics of the reflected signal obtained in step 5, including the reflection amplitude, phase change and pulse width, etc.; the specific steps are as follows

[0084] According to the change of the reflection coefficient measured under different signal frequencies in step 5, the frequency response curve is established; the reflection power coefficient R = |Γ| 2 is obtained from the reflection coefficient, and the fault type and switch state are judged by the reflection coefficient and the reflection power coefficient.

[0085] The fault type and switch state are judged by the reflection coefficient and the reflection power coefficient, and the specific steps are as follows:

[0086]

[0087] A high-frequency signal detection system for grounding cables, comprising

[0088] cable reflection signal acquisition module, used for capturing the partial reflection signal of the detection signal injected into the cable at the discontinuous point in the cable in step 1;

[0089] acquisition signal sparse representation module, used for sparse transformation of the reflection signal captured in step 1 in step 2, extracting key features in the signal and removing redundant information in the signal to obtain a sparse part containing signal features;

[0090] compressed sampling module, used for constructing a Gaussian measurement matrix in step 3, randomly projecting the sparse signal obtained in step 2 to realize compressed sampling of the sparse signal, and outputting the compressed sparse signal;

[0091] signal reconstruction module, used for outputting the reconstructed original signal containing cable fault and positioning information by sparse signal reconstruction algorithm in step 3 in step 4;

[0092] fault detection and positioning module, used for obtaining the characteristics of the reflection signal by performing spectral analysis on the reconstructed original signal output in step 4 in step 5;

[0093] fault type and switch state judgment module, used for identifying the fault signal and judging the cable fault type and the cable switch state and switch operation according to the characteristics of the reflection signal obtained in step 5 in step 6.

[0094] Table 1 Advantage of the invented detection method over the traditional method for cables of different lengths

[0095] .

Claims

1. A method for detecting high-frequency signals in a grounding cable, characterized in that, Specifically, the following steps are included: Step 1: Inject a detection signal into the cable to generate partial reflected signals at discontinuities in the cable. Capture the reflected signals using a cable reflection signal acquisition module. Step 2: Perform sparse transformation on the reflected signal captured by the cable reflection signal acquisition module to extract key features from the signal, including the time difference, amplitude change, attenuation coefficient and frequency response of the reflected signal, which can reflect the cable status or fault location. Remove redundant information from the signal to provide the sparse part containing the signal features, i.e., the sparse signal, for the compression sampling module. Step 3: Construct a Gaussian measurement matrix, perform random projection on the sparse signal provided in Step 2 to achieve compressed sampling of the sparse signal, and output the compressed sparse signal. Step 4: Compare the compressed signal y and wavelet basis matrix obtained in Step 3. The compressed signal y is defined as the original signal that has been compressed and sampled using the Gaussian measurement matrix Ф. The residual signal r0 is initialized as the compressed signal y, r0 = y, and the residual is set to the original observation signal. The index of the wavelet basis vector is selected. Selecting the current residual r from the wavelet basis matrix A k The most relevant basis vectors are as follows: 1) Select the basis vector with the highest correlation: find the one that maximizes the correlation. Right now index i k A i It is the i-th column of the wavelet basis matrix A, representing the i-th wavelet basis vector; 2) Index the selected wavelet basis vectors Add to index collection S k Chinese: S k+1 =S k ∪{i k }; 3) Use the least squares method in the current index set S k+1 The optimal sparse coefficients of the reconstructed signal are calculated above, i.e. in, The submatrix formed by the selected basis vectors; 4) Update residuals 5) Set the residual threshold ε1, and the upper limit of the number of iterations is K; 6) When the residual r k+1 The algorithm stops when the number of iterations is less than the preset threshold ε1, or when the number of iterations reaches K. 7) Output the reconstructed original signal containing cable fault and location information. Step 5: Reconstruct the original signal output from Step 4. Perform spectral analysis to obtain the characteristics of the reflected signal; Step 6: Based on the characteristics of the reflected signal obtained in Step 5, including reflection amplitude, phase change, and pulse width, establish a frequency response curve by measuring the change in reflection coefficient at different signal frequencies; derive the formula for calculating the reflection power coefficient R = |Γ| from the reflection coefficient. 2 The fault signal is identified by the reflection coefficient and reflection power coefficient, and the cable fault type, cable switch status and switch operation are judged: the fault type includes: short circuit fault, open circuit fault and ground fault; the switch status includes: switch closed and switch open.

2. The method for detecting high-frequency signals in a grounding cable according to claim 1, characterized in that, The specific method for step 2 is as follows: The reflected signal captured by the cable reflection signal acquisition module in step 1 is denoted as a sparse signal. For sparse signals The Discrete Wavelet Transform (DWT) is applied to represent a sparse structure θ = DWT(x), where θ represents the sparse coefficients under the wavelet basis. The matrix representation of the DWT is the wavelet basis matrix, composed of wavelet basis functions. The wavelet basis functions include the mother wavelet function ψ. j,k (t)=2 j / 2 ψ(2 j tk) and scaling function Where j represents the scale and k represents the translation; for a cable reflection signal x of length N, the discrete wavelet transform yields the wavelet basis matrix as follows: Let ψ0(t) be an N×N matrix. k ) represents the mother wavelet function at different scales j and sampling points t. k The value below, This represents the scaling function at different scales j and sampling points t. k The value below; Then, a threshold ε is set, and coefficients smaller than ε are set to zero to obtain the sparsified coefficients. A sparse signal x is represented as x = Aθ, where A is the wavelet basis matrix representing the cable reflection signal, and θ is the sparse coefficient under the wavelet basis.

3. The method for detecting high-frequency signals in a grounding cable according to claim 1, characterized in that, The specific method for step 3 is as follows: Construct a random Gaussian measurement matrix Ф of M×N dimensions, and the wavelet basis matrix is where M << N, and each element φ in the random Gaussian measurement matrix Ф ij is a random variable, following a Gaussian distribution with a mean of 0 and a variance of 1: Through the random Gaussian measurement matrix Ф, the sparse signal x in step 2 is randomly projected to obtain the compressed signal y = Фx, reducing the number of sampling points.

4. The method for detecting high-frequency signals in a grounding cable according to claim 1, characterized in that, The specific method for step 5 is as follows: The reconstructed original signal is obtained through step 4. Calculate the reflection coefficient at different injected signal frequencies: 1) Reconstruct the original signal Transform to the frequency domain V(f) using Laplace transform; 2) The transformed frequency domain V(f) includes the injected signal V(f). zhuru and the reflected signal V(f) fans That is, V(f) = V(f) fans +V(f) zhuru Then, the reflection coefficient at different injected signal frequencies is calculated. Among them, V fans (f) is the amplitude of the reflected signal, V zhuru (f) is the amplitude of the injected signal; By injecting different signal frequencies, the characteristics of the reflected signal, namely the spectrum formed by the change of the reflection coefficient with frequency, are obtained.

5. The method for detecting high-frequency signals in a grounding cable according to claim 4, characterized in that, The fault type and switch status are determined by using the reflection coefficient and reflection power coefficient, as detailed below. Fault type judgment: Γ≈-1 indicates a short circuit fault, Γ≈1 indicates an open circuit fault, and Γ∈[-1,0] indicates a ground fault; Switch status judgment: R<0.1 indicates a closed switch, and R>0.8 indicates an open switch.

6. A high-frequency signal detection system for grounding cables, characterized in that, The method for detecting high-frequency signals in a grounding cable as described in any one of claims 1 to 5 specifically includes: The cable reflection signal acquisition module is used to capture the partial reflection signal generated at the discontinuity point in the cable of the detection signal injected into the cable in step 1. The acquisition signal sparse representation module is used in step 2 to perform sparse transformation on the reflected signal captured in step 1, extract key features from the signal, remove redundant information from the signal, and obtain a sparse part containing signal features. The compressed sampling module is used in step 3 to construct a Gaussian measurement matrix, randomly project the sparse signal obtained in step 2, realize the compressed sampling of the sparse signal, and output the compressed sparse signal. The signal reconstruction module is used in step 4 to convert the compressed sparse signal in step 3 into a sparse signal reconstruction algorithm and output a reconstructed original signal containing cable fault and location information. The fault detection and location module is used in step 5 to perform spectrum analysis on the reconstructed original signal output in step 4 to obtain the characteristics of the reflected signal. The fault type and switch status judgment module is used in step 6 to identify the fault signal based on the characteristics of the reflected signal obtained in step 5, and to judge the cable fault type, cable switch status and switch operation.

7. A high-frequency signal detection device for grounding cables, characterized in that, include: A memory and a processor, the memory storing a computer program that causes the processor to execute the high-frequency signal detection method for grounding cables according to any one of claims 1 to 5, based on the stored computer program.

Images