A cable defect positioning method and system based on principal component analysis

By combining principal component analysis and singular value decomposition with Hamming self-convolution window processing, the problems of multiple reflections and noise effects in cable defect localization are solved, achieving higher accuracy in cable defect localization.

CN117192287BActive Publication Date: 2026-11-10GUANGDONG POWER GRID CO LTD +1
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202311171696.X
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2023-09-12
Publication Date
2026-11-10
Estimated Expiration
2043-09-12

AI Technical Summary

Technical Problem

Existing methods such as traveling wave method, frequency domain reflection method and time-frequency domain method have problems with low accuracy in cable defect location, especially due to multiple reflections and noise effects leading to inaccurate location results.

Method used

A principal component analysis-based approach is adopted. By constructing the Hankel matrix from the imaginary part of the frequency domain reflection signal of the cable reflection, principal component analysis and singular value decomposition are performed to eliminate the effects of multiple reflections and noise. The Hamming self-convolution window is used to process the signal to improve the accuracy of FFT calculation. Finally, spectral peak search is performed to locate the defect.

Benefits of technology

It improves the accuracy and efficiency of cable defect location, eliminates the effects of multiple reflections and noise, and ensures the accuracy of the location spectrum.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN117192287B_ABST
    Figure CN117192287B_ABST
Patent Text Reader

Abstract

The application discloses a cable defect positioning method and system based on principal component analysis, which comprises the following steps: inputting a first FDR signal into a cable to be detected, extracting the imaginary part of a second FDR signal reflected by the cable to be detected, and constructing a Hankel matrix according to the imaginary part; performing principal component analysis and singular value decomposition on the Hankel matrix to obtain a plurality of principal components corresponding to the Hankel matrix; performing cross-correlation operation on the plurality of principal components and the imaginary part to obtain a reconstructed signal corresponding to the second FDR signal; performing windowing processing on the reconstructed signal, performing FFT calculation on the reconstructed signal after the windowing processing, obtaining a peak value at a cable defect, performing spectral peak searching according to the peak value, and obtaining defect position information of the cable to be detected, thereby improving the accuracy of cable defect positioning.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of power equipment condition monitoring and fault diagnosis technology, and in particular to a cable defect location method and system based on principal component analysis. Background Technology

[0002] As the service life of power distribution cables increases, insulation aging becomes a prominent issue, such as water treeing in the cable body or moisture absorption at the interface of aging joints. These problems can lead to an abnormal increase in ultra-low frequency dielectric loss. However, ultra-low frequency dielectric loss testing can only determine the overall degree of aging of the cable, not the location of the aging. Replacing the entire cable would be difficult and costly, making the treatment of aging cables uneconomical. Therefore, the detection of cable defects is of great research significance.

[0003] Existing traveling wave methods mainly include: Time Domain Reflectometry (TDR) for detecting cable defects, which obtains defect information by injecting pulse signals into the cable and analyzing the time delay estimates of the input and output signals; Frequency Domain Reflectometry (FDR), which injects frequency-sweeping signals into the cable, containing more high-frequency content and more information, and is more accurate in locating weak defects; and Time-Frequency Domain Reflectometry (TFDR), which combines the characteristics of TDR and FDR, obtains the location spectrum by injecting a Gaussian envelope signal into the cable and performing Wigner time-frequency analysis on the reflected signal.

[0004] However, the TDR method has a narrow input signal frequency range, thus failing to fully reflect the frequency domain information of cable defects. Furthermore, peaks from multiple reflections can easily appear in the location spectrum, affecting the judgment of defect location results. Since the detection signal at the cable head is formed by the superposition of signals after multiple refractions, the detection signal includes frequencies other than the inherent frequency of the defect location, leading to false peaks in the FDR detection results. This makes the FDR location effect less than ideal. Additionally, because the Wigner distribution does not satisfy the principle of double superposition, multiple reflections can easily cause cross-phase in the TFDR location spectrum, generating false peaks and affecting the location results, resulting in low accuracy in cable defect location. Summary of the Invention

[0005] To address the aforementioned technical problems, this invention discloses a cable defect localization method and system based on principal component analysis, which improves the accuracy and efficiency of cable defect localization.

[0006] To achieve the above objectives, in a first aspect, the present invention discloses a cable defect localization method based on principal component analysis, comprising:

[0007] A first FDR signal is input to the cable under test, and the imaginary part of the second FDR signal reflected by the cable under test is extracted. The Hankel matrix is ​​then constructed based on the imaginary part.

[0008] Principal component analysis and singular value decomposition are performed on the Hankel matrix to obtain several principal components corresponding to the Hankel matrix;

[0009] The principal components and the imaginary part are cross-correlated to obtain the reconstructed signal corresponding to the second FDR signal;

[0010] The reconstructed signal is windowed, and the windowed reconstructed signal is subjected to FFT calculation to obtain the peak value at the cable defect. Based on the peak value, a spectral peak search is performed to obtain the defect location information of the cable to be tested.

[0011] This invention discloses a cable defect localization method based on principal component analysis. First, the cable to be tested emits a first FDR signal, and the imaginary part of the second FDR signal reflected by the cable is extracted. This allows for subsequent defect localization based on the rich frequency domain information inherent in the imaginary part. After obtaining the imaginary part, principal component analysis and singular value analysis are performed to obtain several principal components. Subsequently, cross-correlation operations are performed based on these principal components and the imaginary part to eliminate the influence of multiple reflections and noise, ensuring the localization spectrum is unaffected by multiple reflections and improving the accuracy and efficiency of subsequent defect localization. After obtaining the reconstructed signal, windowing is applied to overcome the spectral leakage problem caused by subsequent FFT operations on the reconstructed signal, making the localization spectrum more accurate. After windowing, FFT calculation is performed on the reconstructed signal, and the peak value corresponding to the defect is obtained based on the calculation result. Then, a spectral peak search is performed based on the peak value to locate the defect in the cable to be tested.

[0012] As a preferred example, the process of extracting the imaginary part of the second FDR signal reflected by the cable to be detected and constructing the Hankel matrix based on the imaginary part includes:

[0013] The reflection coefficient spectrum of the second FDR signal is obtained by using a preset frequency sweep signal, and the imaginary part of the second FDR signal is extracted based on the reflection coefficient spectrum.

[0014] Construct a Hankel matrix based on the imaginary part to obtain the Hankel matrix corresponding to the imaginary part.

[0015] This invention uses a frequency sweep method to obtain the reflection coefficient of the cable, and then extracts the corresponding imaginary part through the reflection coefficient, so that the imaginary part has rich frequency domain information, improving the accuracy of subsequent defect location. Furthermore, in order to perform singular value decomposition later, the Hankel matrix is ​​used to both retain all the details of the original data and expand it to a higher dimension, so that it has more feature points. Therefore, the Hankel matrix corresponding to the imaginary part is selected to improve the accuracy of defect location.

[0016] As a preferred example, the process of performing principal component analysis and singular value decomposition on the Hankel matrix to obtain several principal components corresponding to the Hankel matrix includes:

[0017] Principal component analysis is performed on the Hankel matrix to obtain several initial principal components contained in the Hankel matrix;

[0018] Perform singular value decomposition on the Hankel matrix to obtain the number of singular values ​​corresponding to the Hankel matrix;

[0019] Based on the number of singular values, the principal components are obtained by screening from the initial principal components.

[0020] To eliminate the effects of multiple reflections and noise and improve the accuracy of the positioning spectrum, this invention performs principal component analysis on the matrix corresponding to the imaginary part of the reflected signal. At the same time, to improve the positioning efficiency, singular value decomposition is performed so that the principal component data can be reduced in dimensionality based on the number of singular values ​​while preserving data features, thereby reducing the amount of data processing.

[0021] As a preferred example, the step of windowing the reconstructed signal and performing FFT calculation on the windowed reconstructed signal to obtain the peak value at the cable defect includes:

[0022] The reconstructed signal is windowed according to a preset Hamming self-convolution window to obtain the windowed reconstructed signal.

[0023] Calculate the FFT of the reconstructed signal after windowing to obtain the peak value corresponding to the defect in the cable to be inspected.

[0024] This invention employs a Hamming self-convolution window to preprocess the reconstructed signal, which greatly overcomes the spectral leakage problem in the FFT process, making the localization spectrum more accurate and thus improving the accuracy of defect localization.

[0025] Secondly, the present invention also discloses a cable defect location system based on principal component analysis, the system comprising a reflection matrix module, an analysis and solution module, a cross-correlation calculation module, and a defect location module;

[0026] The reflection matrix module is used to input a first FDR signal to the cable under test, extract the imaginary part of the second FDR signal reflected by the cable under test, and construct a Hankel matrix based on the imaginary part.

[0027] The analysis and solution module is used to perform principal component analysis and singular value decomposition on the Hankel matrix to obtain several principal components corresponding to the Hankel matrix;

[0028] The cross-correlation operation module is used to perform cross-correlation operation between the principal components and the imaginary part to obtain the reconstructed signal corresponding to the second FDR signal;

[0029] The defect location module is used to window the reconstructed signal, perform FFT calculation on the windowed reconstructed signal to obtain the peak value at the cable defect, and perform spectral peak search based on the peak value to obtain the defect location information of the cable to be inspected.

[0030] This invention discloses a cable defect location system based on principal component analysis. First, the cable to be inspected emits a first FDR signal, and the imaginary part of the second FDR signal reflected by the cable is extracted. This allows for subsequent defect location based on the rich frequency domain information inherent in the imaginary part. After obtaining the imaginary part, principal component analysis and singular value analysis are performed to obtain several principal components. Subsequently, cross-correlation operations are performed based on these principal components and the imaginary part to eliminate the influence of multiple reflections and noise, ensuring the location spectrum is unaffected by multiple reflections and improving the accuracy and efficiency of subsequent defect location. After obtaining the reconstructed signal, windowing is applied to overcome the spectral leakage problem caused by subsequent FFT operations on the reconstructed signal, making the location spectrum more accurate. After windowing, FFT calculation is performed on the reconstructed signal. The peak value corresponding to the defect is obtained based on the calculation result, and then a spectral peak search is performed based on the peak value to locate the defect in the cable to be inspected.

[0031] As a preferred example, the reflection matrix module includes a reflection extraction unit and a matrix construction unit;

[0032] The reflection extraction unit is used to obtain the reflection coefficient spectrum of the second FDR signal through a preset frequency sweep signal, and extract the imaginary part of the second FDR signal according to the reflection coefficient spectrum;

[0033] The matrix construction unit is used to construct a Hankel matrix based on the imaginary part, thereby obtaining the Hankel matrix corresponding to the imaginary part.

[0034] This invention uses a frequency sweep method to obtain the reflection coefficient of the cable, and then extracts the corresponding imaginary part through the reflection coefficient, so that the imaginary part has rich frequency domain information, improving the accuracy of subsequent defect location. Furthermore, in order to perform singular value decomposition later, the Hankel matrix is ​​used to both retain all the details of the original data and expand it to a higher dimension, so that it has more feature points. Therefore, the Hankel matrix corresponding to the imaginary part is selected to improve the accuracy of defect location.

[0035] As a preferred example, the analysis and solution module includes a principal component analysis unit and a principal component screening unit;

[0036] The principal component analysis unit is used to perform principal component analysis on the Hankel matrix to obtain several initial principal components contained in the Hankel matrix;

[0037] The principal component screening unit is used to perform singular value decomposition on the Hankel matrix to obtain the number of singular values ​​corresponding to the Hankel matrix; and to screen the principal components from the initial principal components based on the number of singular values.

[0038] To eliminate the effects of multiple reflections and noise and improve the accuracy of the positioning spectrum, this invention performs principal component analysis on the matrix corresponding to the imaginary part of the reflected signal. At the same time, to improve the positioning efficiency, singular value decomposition is performed so that the principal component data can be reduced in dimensionality based on the number of singular values ​​while preserving data features, thereby reducing the amount of data processing.

[0039] As a preferred example, the defect location module includes a windowing processing unit and a defect location unit;

[0040] The windowing processing unit is used to perform windowing processing on the reconstructed signal according to a preset Hamming self-convolution window to obtain the windowed reconstructed signal;

[0041] The defect location unit is used to calculate the FFT of the reconstructed signal after windowing processing to obtain the peak value corresponding to the defect in the cable to be inspected.

[0042] This invention employs a Hamming self-convolution window to preprocess the reconstructed signal, which greatly overcomes the spectral leakage problem in the FFT process, making the localization spectrum more accurate and thus improving the accuracy of defect localization.

[0043] Thirdly, the present invention also discloses an electronic device, including a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other through the communication bus; the memory is used to store computer programs; and the processor, when executing the program stored in the memory, implements the cable defect location method based on principal component analysis as described in the first aspect.

[0044] Fourthly, the present invention also discloses a computer-readable storage medium storing a computer program, which, when executed by a processor, implements a cable defect location method based on principal component analysis as described in the first aspect. Attached Figure Description

[0045] Figure 1 This is a flowchart illustrating a cable defect localization method based on principal component analysis disclosed in an embodiment of the present invention.

[0046] Figure 2 This is a schematic diagram of a cable defect location system based on principal component analysis disclosed in an embodiment of the present invention.

[0047] Figure 3 : This is a schematic diagram of a cable model disclosed in an embodiment of the present invention;

[0048] Figure 4 This is a schematic diagram of a cable defect location spectrum obtained by a principal component analysis-based cable defect location method disclosed in an embodiment of the present invention.

[0049] Figure 5 This is a schematic diagram of a location spectrum obtained by defect localization based on the FDR method disclosed in an embodiment of the present invention. Detailed Implementation

[0050] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0051] Example

[0052] This invention discloses a cable defect location method based on principal component analysis. The specific implementation process of the location method can be found in [reference needed]. Figure 1 It mainly includes steps 101 to 104, the steps being:

[0053] Step 101: Input the first FDR signal to the cable under test, extract the imaginary part of the second FDR signal reflected by the cable under test, and construct the Hankel matrix based on the imaginary part.

[0054] In this embodiment, the step mainly includes: obtaining the reflection coefficient spectrum of the second FDR signal through a preset frequency sweep signal, extracting the imaginary part of the second FDR signal based on the reflection coefficient spectrum, constructing a Hankel matrix based on the imaginary part, and obtaining the Hankel matrix corresponding to the imaginary part.

[0055] Furthermore, in one possible implementation method provided in this embodiment, firstly, a first FDR signal is input into the cable to be tested, and a second FDR signal reflected back by the cable to be tested according to the first FDR signal is obtained. The second FDR signal is then swept using a preset frequency sweep signal to obtain the reflection coefficient spectrum of the cable. The expression for the reflection coefficient spectrum is:

[0056]

[0057] Based on the reflection coefficient spectrum, the imaginary part of the second FDR signal is extracted, and the expression for the imaginary part is:

[0058]

[0059] After obtaining the imaginary part of the reflected signal, a Hankel matrix, denoted as H, is constructed using the imaginary part r(ω).

[0060]

[0061] In this embodiment, the reflection coefficient of the cable is obtained by frequency sweeping, and then the corresponding imaginary part is extracted through the reflection coefficient, so that the imaginary part has rich frequency domain information, which improves the accuracy of subsequent defect location. Furthermore, in order to perform singular value decomposition later, the Hankel matrix is ​​used to generate the imaginary part, which not only retains all the details of the original data, but also expands it to a higher dimension, so that it has more feature points. This is to improve the accuracy of defect location.

[0062] Step 102: Perform principal component analysis and singular value decomposition on the Hankel matrix to obtain several principal components corresponding to the Hankel matrix.

[0063] In this embodiment, the step mainly includes: performing principal component analysis on the Hankel matrix to obtain several initial principal components contained in the Hankel matrix; performing singular value decomposition on the Hankel matrix to obtain the number of singular values ​​corresponding to the Hankel matrix; and selecting the several principal components from the several initial principal components based on the number of singular values.

[0064] Furthermore, in one possible implementation method provided in this embodiment, the Hankel matrix is ​​decomposed using Principal Component Analysis (PCA) and Singular Value Decomposition (SVD), respectively. The expressions for the Principal Component Analysis and Singular Value Decomposition are as follows:

[0065]

[0066] In the formula, W is HH T The eigenvalue matrix, ∑1 is HH T The diagonal matrix formed by the eigenvalues ​​of H is given by: U is the left singular matrix of H, ∑2 is the diagonal matrix formed by the singular values ​​of H, and V is the right singular matrix of H. Then, based on the number q singular values ​​in ∑2, the first q eigenvalues ​​of W are retained to obtain a new eigenvalue matrix W'. Dimensionality reduction is then applied to H, resulting in the following matrix composed of several principal components:

[0067] H′=W′H

[0068] In this embodiment, in order to eliminate the influence of multiple reflections and noise and improve the accuracy of the positioning spectrum, this step performs principal component analysis on the matrix corresponding to the imaginary part of the reflected signal. At the same time, in order to improve the positioning efficiency, singular value decomposition is performed so that the principal component data can be reduced in dimensionality based on the number of singular values ​​while preserving data features, thereby reducing the amount of data processing.

[0069] Step 103: Perform cross-correlation operation between the principal components and the imaginary part to obtain the reconstructed signal corresponding to the second FDR signal.

[0070] In this embodiment, the step is as follows: extract the first N values ​​from the matrix H' composed of the plurality of principal components to form a principal component sequence s(ω), and perform cross-correlation between s(ω) and r(ω), wherein the expression for the cross-correlation operation is as follows:

[0071]

[0072] In the formula, * denotes complex conjugation.

[0073] Step 104: Window the reconstructed signal and perform FFT calculation on the windowed reconstructed signal to obtain the peak value at the cable defect. Perform spectral peak search based on the peak value to obtain the defect location information of the cable to be tested.

[0074] In this embodiment, the step mainly includes: windowing the reconstructed signal according to a preset Hamming self-convolution window to obtain the windowed reconstructed signal; calculating the FFT of the windowed reconstructed signal to obtain the peak value corresponding to the defect in the cable to be detected.

[0075] Furthermore, in one possible implementation of this embodiment, in order to reduce spectral leakage during the FFT process, a Hamming self-convolution window needs to be added to R(k). The expression for adding the Hamming self-convolution window is as follows:

[0076]

[0077] Finally, calculate The FFT is used to obtain the peak value corresponding to the defect in the detected cable, and the expression for the peak value is as follows:

[0078]

[0079] In this embodiment, the Hamming self-convolution window is used to preprocess the reconstructed signal, which greatly overcomes the spectral leakage problem in the FFT process, making the localization spectrum more accurate and thus improving the accuracy of defect localization.

[0080] Specifically, the cable defect location method based on principal component analysis provided in this invention and the traditional FDR method are used to locate defects in the cable under test. The length of the cable under test is set to l = 600m, the defect location x1 = 200m, and the defect length d1 = 0.5m. A 0.15-100MHz sinusoidal sweep frequency signal is input at the beginning of the cable, and 3000 samples are taken to locate the defect. The cable model of the cable under test can be referenced... Figure 3 The location spectrum obtained by defect localization using the method provided in this embodiment of the invention can be referred to... Figure 4 The location spectrum obtained by the traditional FDR method for defect localization can be referenced. Figure 5 ,Depend on Figure 4 and Figure 5 As can be seen from the comparison, the accuracy of defect location in the embodiments of the present invention is higher than that of the traditional FDR method.

[0081] On the other hand, the present invention also discloses a cable defect location system based on principal component analysis, the specific structural composition of which can be referred to Figure 2 The system includes a reflection matrix module 201, an analysis and solution module 202, a cross-correlation calculation module 203, and a defect location module 204.

[0082] The reflection matrix module 201 is used to input a first FDR signal to the cable under test, extract the imaginary part of the second FDR signal reflected by the cable under test, and construct a Hankel matrix based on the imaginary part.

[0083] The analysis and solution module 202 is used to perform principal component analysis and singular value decomposition on the Hankel matrix to obtain several principal components corresponding to the Hankel matrix.

[0084] The cross-correlation operation module 203 is used to perform cross-correlation operation between the plurality of principal components and the imaginary part to obtain the reconstructed signal corresponding to the second FDR signal.

[0085] The defect location module 204 is used to perform windowing processing on the reconstructed signal, and to perform FFT calculation on the windowed reconstructed signal to obtain the peak value at the cable defect. Based on the peak value, a spectral peak search is performed to obtain the defect location information of the cable to be inspected.

[0086] In this embodiment, the reflection matrix module 201 includes a reflection extraction unit and a matrix construction unit.

[0087] The reflection extraction unit is used to obtain the reflection coefficient spectrum of the second FDR signal through a preset frequency sweep signal, and extract the imaginary part of the second FDR signal based on the reflection coefficient spectrum.

[0088] The matrix construction unit is used to construct a Hankel matrix based on the imaginary part, thereby obtaining the Hankel matrix corresponding to the imaginary part.

[0089] In this embodiment, the analysis and solution module 202 includes a principal component analysis unit and a principal component screening unit.

[0090] The principal component analysis unit is used to perform principal component analysis on the Hankel matrix to obtain several initial principal components contained in the Hankel matrix.

[0091] The principal component screening unit is used to perform singular value decomposition on the Hankel matrix to obtain the number of singular values ​​corresponding to the Hankel matrix; and to screen the principal components from the initial principal components based on the number of singular values.

[0092] In this embodiment, the defect location module 204 includes a windowing processing unit and a defect location unit.

[0093] The windowing processing unit is used to perform windowing processing on the reconstructed signal according to a preset Hamming self-convolution window to obtain the windowed reconstructed signal.

[0094] The defect location unit is used to calculate the FFT of the reconstructed signal after windowing processing to obtain the peak value corresponding to the defect in the cable to be inspected.

[0095] In addition to the methods and systems described above, embodiments of the present invention also provide an electronic device and a computer-readable storage medium. The electronic device includes a processor, a communication interface, a memory, and a communication bus. The processor, communication interface, and memory communicate with each other via the communication bus. The memory stores a computer program. When the processor executes the program stored in the memory, it implements the cable defect location method based on principal component analysis described in the embodiments of the present invention. The computer-readable storage medium stores a computer program, which, when executed by the processor, implements the cable defect location method based on principal component analysis described in the embodiments of the present invention.

[0096] This invention provides a cable defect location method and system based on principal component analysis (PCA). It extracts the principal components from the modal aliasing signal at the cable's head end and performs cross-correlation with the original signal to eliminate the influence of additional frequencies and noise caused by multiple refractions, thus eliminating the impact of spurious peaks on the location results and improving the accuracy of defect location. Compared to TDR and TFDR, this technology uses a frequency sweep method to obtain the transfer function of the submarine cable, possessing rich frequency domain information. Compared to traditional FFT-based FDR, this technology uses PCA to extract the first reflected wave of the submarine cable's transfer function for reconstruction and cross-correlation, eliminating the influence of multiple reflections and noise, making the location spectrum unaffected by multiple reflections. Furthermore, this embodiment uses a 4th-order Hamming self-convolution window to preprocess the reconstructed signal, greatly overcoming the spectral leakage problem in the FFT process and making the location spectrum more accurate.

[0097] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above descriptions are merely specific embodiments of the present invention and are not intended to limit the scope of protection of the present invention. In particular, it should be noted that any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention for those skilled in the art.

Claims

1. A method for locating cable defects based on principal component analysis, characterized in that, include: A first FDR signal is input to the cable under test, and the imaginary part of the second FDR signal reflected by the cable under test is extracted. The Hankel matrix is ​​then constructed based on the imaginary part. Principal component analysis and singular value decomposition are performed on the Hankel matrix to obtain several principal components corresponding to the Hankel matrix; The principal components and the imaginary part are cross-correlated to obtain the reconstructed signal corresponding to the second FDR signal; The reconstructed signal is windowed, and the windowed reconstructed signal is subjected to FFT calculation to obtain the peak value at the cable defect. Based on the peak value, a spectral peak search is performed to obtain the defect location information of the cable to be tested.

2. The cable defect localization method based on principal component analysis as described in claim 1, characterized in that, The step of extracting the imaginary part of the second FDR signal reflected by the cable under test and constructing the Hankel matrix based on the imaginary part includes: The reflection coefficient spectrum of the second FDR signal is obtained by using a preset frequency sweep signal, and the imaginary part of the second FDR signal is extracted based on the reflection coefficient spectrum. Construct a Hankel matrix based on the imaginary part to obtain the Hankel matrix corresponding to the imaginary part.

3. The cable defect localization method based on principal component analysis as described in claim 1, characterized in that, The process of performing principal component analysis and singular value decomposition on the Hankel matrix to obtain several principal components corresponding to the Hankel matrix includes: Principal component analysis is performed on the Hankel matrix to obtain several initial principal components contained in the Hankel matrix; Perform singular value decomposition on the Hankel matrix to obtain the number of singular values ​​corresponding to the Hankel matrix; Based on the number of singular values, the principal components are obtained by screening from the initial principal components.

4. The cable defect localization method based on principal component analysis as described in claim 1, characterized in that, The step of windowing the reconstructed signal and performing FFT calculation on the windowed reconstructed signal to obtain the peak value at the cable defect includes: The reconstructed signal is windowed according to a preset Hamming self-convolution window to obtain the windowed reconstructed signal. Calculate the FFT of the reconstructed signal after windowing to obtain the peak value corresponding to the defect in the cable to be inspected.

5. A cable defect location system based on principal component analysis, characterized in that, The system includes a reflection matrix module, an analysis and solution module, a cross-correlation calculation module, and a defect location module; The reflection matrix module is used to input a first FDR signal to the cable under test, extract the imaginary part of the second FDR signal reflected by the cable under test, and construct a Hankel matrix based on the imaginary part. The analysis and solution module is used to perform principal component analysis and singular value decomposition on the Hankel matrix to obtain several principal components corresponding to the Hankel matrix; The cross-correlation operation module is used to perform cross-correlation operation between the principal components and the imaginary part to obtain the reconstructed signal corresponding to the second FDR signal; The defect location module is used to window the reconstructed signal, perform FFT calculation on the windowed reconstructed signal to obtain the peak value at the cable defect, and perform spectral peak search based on the peak value to obtain the defect location information of the cable to be inspected.

6. The cable defect location system based on principal component analysis as described in claim 5, characterized in that, The reflection matrix module includes a reflection extraction unit and a matrix construction unit; The reflection extraction unit is used to obtain the reflection coefficient spectrum of the second FDR signal through a preset frequency sweep signal, and extract the imaginary part of the second FDR signal according to the reflection coefficient spectrum; The matrix construction unit is used to construct a Hankel matrix based on the imaginary part, thereby obtaining the Hankel matrix corresponding to the imaginary part.

7. The cable defect location system based on principal component analysis as described in claim 5, characterized in that, The analysis and solution module includes a principal component analysis unit and a principal component screening unit; The principal component analysis unit is used to perform principal component analysis on the Hankel matrix to obtain several initial principal components contained in the Hankel matrix; The principal component screening unit is used to perform singular value decomposition on the Hankel matrix to obtain the number of singular values ​​corresponding to the Hankel matrix; and to screen the principal components from the initial principal components based on the number of singular values.

8. A cable defect location system based on principal component analysis as described in claim 5, characterized in that, The defect location module includes a windowing processing unit and a defect location unit; The windowing processing unit is used to perform windowing processing on the reconstructed signal according to a preset Hamming self-convolution window to obtain the windowed reconstructed signal; The defect location unit is used to calculate the FFT of the reconstructed signal after windowing processing to obtain the peak value corresponding to the defect in the cable to be inspected.

9. An electronic device, characterized in that, It includes a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other through the communication bus; the memory is used to store computer programs; and the processor, when executing the program stored in the memory, implements the cable defect location method based on principal component analysis as described in any one of claims 1-4.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, which, when executed by a processor, implements a cable defect location method based on principal component analysis as described in any one of claims 1-4.

Citation Information

Patent Citations

  • Singular decomposition-based signal noise reduction method

    CN110619296A

  • Power grid phasor mobile measurement method and system based on improved matrix pencil, and medium

    CN115219787A