Cable fault positioning method and system based on reflected wave decomposition and kurtosis sorting

Abstract

The invention discloses a cable fault positioning method and system based on reflected wave decomposition and kurtosis sorting, and aims to solve the problem that a cable fault is difficult to detect under a low impedance change rate. The method comprises the steps of transmitting an incident excitation signal to a cable to obtain a first reflection signal in the cable; an improved VMD algorithm is adopted for decomposition to obtain K IMF components, and the IMF with the maximum kurtosis is selected as a second reflection signal; acquiring a time-frequency distribution curve of the second reflected signal by using short-time Fourier transform and frequency spectrum redistribution technologies; and constructing a two-dimensional energy-time curve by extracting an energy value and time under the frequency of the maximum energy amplitude point, and determining a fault position. The system is composed of an incident excitation module, a decomposition module, an STFT-RSP module and a fault positioning output module, and the positioning method is achieved. According to the method, the limitation of a traditional spread spectrum time domain reflection positioning method in cable fault detection is effectively overcome, and an efficient and accurate instructive technology is provided for cable fault positioning and detection.

Description

Technical Field

[0001] The invention belongs to the field of power cable fault detection and positioning, and in particular relates to a cable fault positioning method and system for reflected wave decomposition and kurtosis sorting. Background Art

[0002] Power cables are usually laid in hidden areas such as distribution networks. Due to the complex wiring harness, it is difficult to visually or physically check the health of the cables and prevent cable failures. Therefore, it is of great significance to develop a convenient, high-precision, and non-destructive cable fault detection method.

[0003] Existing cable fault detection methods include bridge method, impulse flashover method, reflection method, etc. Among them, the bridge method is a two-terminal detection method. Although it can accurately locate the fault distance of single-phase grounding, relative short circuit or short circuit grounding of power cables, it requires more auxiliary equipment and complex professional test data analysis methods. The method is complex and not easy to form a portable test device. The reflection method is to inject the excitation signal into the cable under test, locate the fault, and determine the fault type according to the waveform of the reflected signal. It has the advantages of non-destructive and intuitive, and has been widely used in soil moisture detection, PCB fault detection, cable fault detection and other fields. However, time domain reflection technology is sensitive to noise interference and has limited detection capabilities for non-obvious faults (such as weak signal attenuation, intermittent faults, etc.). In addition, although the correlation analysis between the reflected signal and the incident signal can effectively identify the fault type, in time-frequency reflection analysis, the traditional Wigner-Ville distribution (WVD) method often interferes with the accuracy of the detection results due to the introduction of cross terms.

[0004] To address the above issues, especially the challenges of signal attenuation, noise interference, and reduced reflected signal amplitude caused by the small impedance change rate when a cable fails, an innovative cable fault location method is needed. Summary of the invention

[0005] In order to overcome the technical problem that the fault signal captured by the reflection method in the above-mentioned prior art has nonlinear and non-smooth characteristics and it is difficult to extract the fault signal under the interference of external noise, the present invention adopts improved variational mode decomposition (VMD) to decompose the SSTDR results, and selects the most relevant components to obtain time-frequency time difference information and fault location information.

[0006] The technical solution of the present invention is as follows:

[0007] In a first aspect, a cable fault location method for reflected wave decomposition and kurtosis sorting is characterized in that it includes:

[0008] S1. transmitting an incident excitation signal to the cable and receiving a first reflected signal inside the cable;

[0009] S2. By comparing the kurtosis values ​​of the IMF components, the IMF component with the largest kurtosis value is selected as the second reflection signal inside the cable;

[0010] S3. Apply short-time Fourier transform and spectrum redistribution technology to the second reflected signal to obtain a time-frequency distribution curve of the second reflected signal;

[0011] S4. Select the frequency where the maximum energy amplitude point is located in the time-frequency distribution curve of the second reflected signal, extract the energy value and time information at this frequency, construct a two-dimensional energy-time curve, and determine the fault location of the cable according to the maximum energy peak time difference in the two-dimensional energy-time curve and the known wave speed in the cable.

[0012] Furthermore, the incident excitation signal is generated by binary phase shift keying modulation of a pseudo noise (PN) code and a sine wave, and its formula is:

[0013] S in (t) = y(t)cos(2πf sin t)

[0014] Among them, f sin is the frequency of the sine wave, t is the time point, y(t) can be expressed as:

[0015]

[0016] Where g(t) is the width of T c A rectangular pulse, a n The values ​​-1 and 1 correspond to the binary symbols 0 and 1 respectively. c is the chip duration of the PN code, and n is the number of bits of the PN code.

[0017] Furthermore, the sine wave frequency f sin The value range is [30MHz, 80MHz], and the chip duration of the PN code is T c The value range of is [0.01μs, 0.1μs], and the value range of n is [23,39].

[0018] Furthermore, the improved VMD algorithm specifically includes:

[0019] 1) Define the first reflected signal f(t) inside the cable as

[0020] f(t)=αs ref (t)+n(t)

[0021] Among them, s ref (t) is the actual reflected signal, n(t) is the noise, and α is the weight.

[0022] 2) The Lagrangian multiplication operator λ and the quadratic penalty factor β are introduced to transform the constrained variational VMD decomposition problem into an unconstrained variational problem. The following augmented form of the Lagrangian expression is obtained.

[0023]

[0024] Where k is the number of IMFs, uk and wk are the kth IMF after decomposition and its corresponding central frequency, δ(t) is the Dirac function, λ is the Lagrangian factor, and L is the decomposed IMF set.

[0025] 3) The center frequency comparison method is used to determine the optimal value of k.

[0026] 4) Compare the kurtosis of k IMFs and determine the one with the largest kurtosis value as the second reflected signal inside the cable.

[0027] Furthermore, in the cable fault location method of reflected wave decomposition and kurtosis sorting, in the step (2), the center frequency comparison method is used to determine the optimal value of k, which is characterized by specifically including.

[0028] 1) Determine VMD parameters (preset number of IMFs, bandwidth constraint / penalty factor, noise tolerance, convergence criteria, etc.).

[0029] 2) Perform VMD decomposition on the first reflected signal inside the cable to obtain the IMF component.

[0030] 3) Use Fast Fourier Transform (FFT) to calculate the center frequency of each IMF.

[0031] 4) Determine whether there is aliasing at the center frequency of each IMF.

[0032] 5) If there is no cross term at the center frequency, the k value is increased by 1 and the process returns to step 2). When there is a cross term at the center frequency, the decomposition stops and k-1 is the optimal IMF value.

[0033] Furthermore, the short-time Fourier transform and spectrum redistribution technology is used to obtain the time-frequency distribution curve of the second reflected signal, and the specific content is as follows.

[0034] Often, the amplitude of the reflection peak of the second reflection signal inside the cable caused by the fault is not obvious, and the signal positioning is insufficient only by analyzing the frequency domain.

[0035] 1) For this purpose, the second reflected signal inside the cable is x(t), and its short-time Fourier transform STFT expression is:

[0036]

[0037] E(t,f;h)=10log|STFT(t,f;h)| 2

[0038] Among them, h(τ-t) is the window function, that is, a Gaussian window with a length equal to 65 steps, E(t,f;h) is the energy value, and t and f are time and frequency scales respectively.

[0039] 2) The spectrum redistribution technology of the STFT time-frequency diagram of the second reflected signal inside the spectrum cable is implemented according to the following formula.

[0040]

[0041] Among them, * is the conjugate transpose, δ() is the activation function, RSP x (t′,f′) is the result of spectrum reallocation.

[0042] Further, the determining of the cable fault location specifically includes:

[0043] The location of the fault point or terminal open circuit point is determined by measuring and calculating the maximum energy peak time difference in the two-dimensional energy-time curve and combining it with the known wave speed in the cable.

[0044] In a second aspect, a cable fault location system for reflected wave decomposition and kurtosis sorting is characterized by comprising:

[0045] Incident excitation module: used to transmit an incident excitation signal to the cable and obtain the first reflected signal inside the cable;

[0046] Decomposition module: It is used to decompose K IMF components of the first reflection signal inside the cable using the improved VMD algorithm, and select the IMF with the maximum kurtosis as the second reflection signal inside the cable;

[0047] STFT-RSP module: It is used to take short-time Fourier transform and spectrum redistribution technology to the second reflected signal inside the cable to obtain the time-frequency distribution curve of the second reflected signal;

[0048] Fault location output module: It is used to select the frequency of the maximum energy amplitude point in the time-frequency distribution curve of the second reflected signal, extract the energy value and time at this frequency, and obtain a two-dimensional energy-time curve to determine the cable fault location.

[0049] The principle of the present invention is as follows:

[0050] The location method of time-frequency domain reflection mainly relies on reflected waves to detect cable faults. When the characteristic impedance of the cable is discontinuous, reflected waves will be formed. When the characteristic impedance at the fault location approaches infinity, it is judged as an open circuit fault, and when the characteristic impedance at the fault location approaches zero, it is judged as a short circuit fault.

[0051] The amplitude of the reflected signal is directly related to the rate of change of the characteristic impedance of the fault point. Therefore, it is necessary to study the relationship between the characteristic impedance of the cable and the degree of damage. The characteristic impedance of the coaxial cable is calculated as follows:

[0052]

[0053] Wherein, Z0 is the unit characteristic impedance; R, L, C, G are the unit distributed resistance, inductance, capacitance, and conductance respectively; ω is the frequency.

[0054] According to the above formula and the cable damage depth and impedance change rate, it can be known that Figure 2 As shown in the figure, when the cable damage degree is low, the coaxial cable impedance change rate is weak. When the damage degree reaches 50%, the characteristic impedance change rate is only about 25%; when the damage degree exceeds 80%, the characteristic impedance change rate increases significantly; when the damage degree reaches 100%, the cable is in an open circuit state, and the characteristic impedance change rate reaches infinity.

[0055] Compared with the prior art, the present invention has the following beneficial effects:

[0056] 1) By using the incident excitation signal generated by binary phase shift keying modulation of pseudo-noise (PN) code and sine wave, the signal's anti-interference ability and penetration are enhanced, making the reflected signal clearer, thereby improving the accuracy of fault location.

[0057] 2) The improved VMD algorithm can more effectively decompose the IMF components in the first reflection signal inside the cable, and select the IMF with the largest kurtosis as the second reflection signal, further reducing noise interference and improving signal quality. The optimal number k of IMF components is determined by the center frequency comparison method, avoiding the problem of over-decomposition or under-decomposition, thereby improving the calculation efficiency and accuracy of the algorithm.

[0058] 3) The combined use of short-time Fourier transform (STFT) and spectrum redistribution technology can capture the time-frequency characteristics of the second reflected signal. Especially when the reflection peak amplitude caused by the fault is not obvious, the spectrum redistribution technology can significantly improve the time-frequency resolution and make the fault characteristics more obvious.

[0059] 4) The method of the present invention is not only applicable to frequency domain analysis, but also combines time domain information, providing strong technical support for the precise location of cable faults. By introducing the Lagrange multiplication operator and the quadratic penalty factor, the constrained variational problem is transformed into an unconstrained problem, further simplifying the calculation process. By constructing a two-dimensional energy-time curve and determining the fault location based on the maximum energy peak time difference and the known cable wave velocity, the automation and intelligence of fault location are realized, human intervention and errors are reduced, and the accuracy and reliability of positioning are improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] Figure 1 It is a flow chart of the cable fault location method of the reflected wave decomposition and kurtosis sorting of the present invention;

[0061] Figure 2 It is a curve chart showing the relationship between cable loss degree and impedance change rate;

[0062] Figure 3 is a flow chart of the improved VMD algorithm of the present invention;

[0063] Figure 4 is the result of the improved VMD decomposition of the present invention;

[0064] Figure 5 It is the time-frequency energy distribution diagram of IMF1 component.

[0065] Figure 6 It is the result of spectrum redistribution of IMF1 component;

[0066] Figure 7 This is a diagram showing the results of the cable fault detection and location by the method of the present invention;

[0067] Figure 8 It is a schematic diagram of a cable fault location system for reflected wave decomposition and kurtosis sorting of the present invention;

[0068] Specific implementation cases

[0069] The technical solution of the present invention is explained in detail below in conjunction with the accompanying drawings and embodiments, but this should not limit the protection scope of the present invention.

[0070] Example 1: Perform fault detection on an actual cable, artificially set the fault position at 3.02 meters and the circuit breaker fault at 10.48 meters.

[0071] A cable fault location method based on reflection wave decomposition and kurtosis sorting is implemented in the following steps.

[0072] (1) transmitting an incident excitation signal to the cable and obtaining a first reflected signal inside the cable;

[0073] (2) The first reflection signal inside the cable is decomposed into K IMF components using the improved VMD algorithm, and the IMF with the largest kurtosis is defined as the second reflection signal inside the cable;

[0074] (3) short-time Fourier transform and spectrum redistribution technology are used to obtain the time-frequency distribution curve of the second reflected signal inside the cable;

[0075] (4) Select the frequency at which the maximum energy amplitude point is located in the time-frequency distribution curve of the second reflected signal, extract the energy value and time at this frequency, and obtain a two-dimensional energy-time curve, thereby determining the cable fault location.

[0076] Specifically, in the step (1), the incident excitation signal has the following specific contents.

[0077] The incident excitation signal is generated by binary phase shift keying modulation of pseudo noise (PN) code and sine wave, and its formula is:

[0078] S in (t) = y(t)cos(2πf sin t)

[0079] Among them, f sin is the frequency of the sine wave, t is the time point, y(t) can be expressed as:

[0080]

[0081] Where g(t) is a rectangular pulse with a width of Tc, the values ​​of an are -1 and 1, corresponding to the binary symbols 0 and 1 respectively, tc is the chip duration of the PN code, and n is the number of bits of the PN code.

[0082] The incident signal after phase delay τ is used as the reference signal, and cross-correlated with the reflected signal f(t) can obtain the kurtosis of the positioning information. Since the delay time τ cannot be less than one chip time, the range of τ can be obtained as:

[0083] t c ≤τ≤nt c

[0084] The effective detection range of the cable under test using the incident excitation signal is g, which can be expressed as:

[0085]

[0086] Wherein, v is the propagation velocity of the incident excitation signal in the measured cable; L and C are the distribution parameters of the measured cable. In this implementation case, the sinusoidal carrier frequency is set to 50MHz, the PN code type is m sequence, the PN code bit number is 31, and the code chip time is 0.02μs.

[0087] Specifically, in the step (2), the improved VMD algorithm specifically includes:

[0088] The VMD algorithm is a signal processing method equivalent to an adaptive filter bank. Its adaptability lies in determining the number of IMF components of given data according to actual conditions, and adaptively matching the optimal center frequency and finite bandwidth of each mode during the search process. VMD can effectively separate the frequency domain of complex signals, and then obtain the effective decomposition components of the given signal, and finally obtain the optimal solution to the variational problem. In order to solve the optimal solution of the variational constraint problem, the Lagrangian multiplication operator λ and the quadratic penalty factor β are introduced to transform the constrained variational problem into an unconstrained variational problem. Eliminate the constraints and obtain the augmented form of the Lagrangian expression. Use the alternating direction multiplication operator method to iteratively update uk, wk and λ, and then iteratively augment the function to finally obtain the optimal solution.

[0089] 1) Define the first reflected signal f(t) inside the cable as

[0090] f(t)=αs ref (t)+n(t)

[0091] Among them, s ref (t) is the actual reflected signal, n(t) is the noise, and α is the weight.

[0092] 2) The Lagrangian multiplication operator λ and the quadratic penalty factor β are introduced to transform the constrained variational VMD decomposition problem into an unconstrained variational problem. The following augmented form of the Lagrangian expression is obtained.

[0093]

[0094] Where k is the number of IMFs, uk and wk are the kth IMF after decomposition and its corresponding central frequency, δ(t) is the Dirac function, λ is the Lagrangian factor, and L is the decomposed IMF set.

[0095] 3) The center frequency comparison method is used to determine the optimal value of k.

[0096] 4) Compare the kurtosis of k IMFs and determine the one with the largest kurtosis value as the second reflected signal inside the cable.

[0097] As attached Figure 3 As shown, the center frequency comparison method is used to determine the optimal value of k, which is characterized by specifically including.

[0098] The decomposition effect of VMD is mainly affected by the number of IMF components. k is the number of IMF components. Conventional decomposition methods will pre-set the number of IMF components, but when the k value is small, the signal decomposition is incomplete and some important components in the signal cannot be obtained; when the k value is too large, the signal will be over-decomposed and the center frequencies of the decomposed signals will be mixed together. Therefore, selecting the most appropriate k value according to the actual situation of the signal is conducive to eliminating the aliasing phenomenon between the decomposed signals and the problem of difficulty in distinguishing close-frequency components.

[0099] Table 1

[0100]

[0101] 1) Determine VMD parameters (preset number of IMFs, bandwidth constraint / penalty factor, noise tolerance, convergence criteria, etc.).

[0102] 2) Perform VMD decomposition on the first reflected signal inside the cable to obtain the IMF component.

[0103] 3) Use Fast Fourier Transform (FFT) to calculate the center frequency of each IMF.

[0104] 4) Determine whether there is aliasing at the center frequency of each IMF.

[0105] 5) If there is no cross term at the center frequency, the k value is increased by 1 and the process returns to step 2). When there is a cross term at the center frequency, the decomposition stops and k-1 is the optimal IMF value.

[0106] Since the main difference between the IMFs is the center frequency, the center frequency comparison method is used to determine the optimal value of k. The main idea of ​​the center frequency comparison analysis is to determine whether there is aliasing in the center frequency of each IMF. The center frequencies of each IMF of the first reflection signal inside the cable in this implementation case under different k are shown in Table 1. When k=6, the center frequency difference between IMF4 and IMF5 is very small, which is prone to frequency aliasing; when k=7, the center frequencies of IMF4 and IMF5 are aliased. Obviously, when the k value continues to increase, the signal will be over-decomposed, and serious aliasing will occur between the center frequencies of each IMF. When k<5, although there is no aliasing, compared with k=5, the first reflection signal inside the cable is not completely decomposed. Therefore, the optimal number of IMFs for the reflection signal in this implementation case is 5. The decomposition results are shown in the attached. Figure 4 As shown, it can be seen that IMF1 has the highest kurtosis. The larger the kurtosis, the more conducive it is to extracting fault features. Therefore, IMF1 is selected for analysis in this implementation case.

[0107] Specifically, in the step (3), short-time Fourier transform and spectrum redistribution technology are used to obtain the time-frequency distribution curve of the second reflected signal, and the specific content is as follows.

[0108] Often, the amplitude of the reflection peak of the second reflection signal inside the cable caused by the fault is not obvious, and the signal positioning is insufficient only by analyzing the frequency domain.

[0109] In order to better locate cable faults, it is an effective method to understand the change of signal spectrum over time. STFT is a signal processing method from the perspectives of time domain and frequency domain. It has good energy concentration and can simultaneously describe the energy intensity of the signal at different times and frequencies. According to the theory of STFT, non-stationary signals are composed of a series of short-time stationary signals. STFT is a Fourier-related transform that multiplies the signal to be transformed by a window function and covers the entire time domain of the signal by translating the window.

[0110] 1) For this purpose, the second reflected signal inside the cable is x(t), and its short-time Fourier transform STFT expression is:

[0111]

[0112] E(t,f;h)=10log|STFT(t,f;h)| 2

[0113] Among them, h(τ-t) is the window function, that is, a Gaussian window with a length equal to 65 steps, E(t,f;h) is the energy value, and t and f are time and frequency scales respectively.

[0114] As attached Figure 5 As shown in the figure, the position of the highest energy intensity point is basically consistent with the position of the second reflected signal inside the cable. However, there are some fuzzy high-energy points between the two points, and the resolution is low. The high-energy points are not concentrated, and the fault cannot be accurately located.

[0115] 2) The spectrum redistribution technology of the STFT time-frequency diagram of the second reflected signal inside the spectrum cable is implemented according to the following formula.

[0116]

[0117] Among them, * is the conjugate transpose, δ() is the activation function, RSP x (t′,f′) is the result of spectrum reallocation.

[0118] As attached Figure 6 As shown, the second reflected signal inside the cable in this implementation case has three significant energy high points, and the corresponding times are approximately from left to right, corresponding to the event signal position, fault point, and disconnection point of the cable terminal, respectively.

[0119] Specifically, determining the cable fault location in (4) includes:

[0120] The location of the fault point or terminal open circuit point is determined by measuring and calculating the maximum energy peak time difference in the two-dimensional energy-time curve and combining it with the known wave speed in the cable.

[0121] As attached Figure 7 As shown, the frequency where the maximum energy amplitude point is located is selected, and the energy value and time at this frequency are extracted to obtain a two-dimensional energy-time curve. Due to the dispersion effect of electromagnetic waves, the propagation characteristics of electromagnetic waves of different frequencies are different. Therefore, there are multiple energy peaks in a period of time. The maximum value in a period of time is selected as the calibration point. According to the wave speed in the cable and the time difference t1 and t2, the fault point position is calculated to be 3.045m, which is very close to the fault point position at 3.02m. The calculated terminal open point position is 13.965m, with an error of 3.4%. According to Figure 7 It can be known from the calculation that the method described can detect and locate the fault of the cable, thus ensuring the accuracy of the detection method.

[0122] Example 2

[0123] like Figure 8 As shown, a cable fault location system for reflected wave decomposition and kurtosis sorting specifically comprises:

[0124] Incident excitation module: used to transmit an incident excitation signal to the cable and obtain the first reflected signal inside the cable;

[0125] Decomposition module: It is used to decompose K IMF components of the first reflection signal inside the cable using the improved VMD algorithm, and select the IMF with the maximum kurtosis as the second reflection signal inside the cable;

[0126] STFT-RSP module: It is used to take short-time Fourier transform and spectrum redistribution technology to the second reflected signal inside the cable to obtain the time-frequency distribution curve of the second reflected signal;

[0127] Fault location output module: It is used to select the frequency of the maximum energy amplitude point in the time-frequency distribution curve of the second reflected signal, extract the energy value and time at this frequency, and obtain a two-dimensional energy-time curve to determine the cable fault location.

[0128] The present invention improves the problem that traditional time-frequency reflection technology cannot effectively detect faults, so that it can more accurately identify faults such as loose connectors and damaged insulation layers. In particular, when the characteristic impedance of the cable changes by more than 0.5 ohms, the method shows good detection effect. By using the improved variable mode decomposition (VMD) method, the method can decompose the reflection signal into multiple intrinsic mode functions (IMFs), and select the IMF with the highest kurtosis for analysis, so as to better extract the fault characteristics, thereby improving the ability to locate the fault. Combined with short-time Fourier transform (STFT) and rearranged spectrum (RSP) technology, the method can provide a clearer time-frequency distribution diagram, which is very useful for understanding the spectral characteristics of the signal changing over time and accurately locating the fault. A cable fault location method and system for reflection wave decomposition and kurtosis sorting in an embodiment of the present invention have significant beneficial effects in improving positioning accuracy, enhancing signal analysis capabilities, optimizing algorithm efficiency, and realizing automation and intelligence, and provide strong technical support for the rapid and accurate location of cable faults.

Claims

1. A cable fault location method based on reflected wave decomposition and kurtosis sorting, characterized in that: include: S1. transmitting an incident excitation signal to the cable and receiving a first reflected signal inside the cable; S2. By comparing the kurtosis values ​​of the IMF components, the IMF component with the largest kurtosis value is selected as the second reflection signal inside the cable; S3. Apply short-time Fourier transform and spectrum redistribution technology to the second reflected signal to obtain a time-frequency distribution curve of the second reflected signal; S4. Select the frequency where the maximum energy amplitude point is located in the time-frequency distribution curve of the second reflected signal, extract the energy value and time information at this frequency, construct a two-dimensional energy-time curve, and determine the fault location of the cable according to the maximum energy peak time difference in the two-dimensional energy-time curve and the known wave speed in the cable.

2. A cable fault location method for reflected wave decomposition and kurtosis sorting according to claim 1, characterized in that: In step S1, the incident excitation signal S in It is generated by binary phase shift keying modulation of pseudo noise (PN) code and sine wave, and the formula is as follows: S in (t)=y(t)cos(2πf sin t) In the formula, t is the time point, f sin is the frequency of the sine wave, y(t) is expressed as: Among them, T c is the chip duration of the PN code, and its value range is [0.01μs, 0.1μs], and g(t) is the width T c A rectangular pulse, a n The values ​​of -1 and 1 correspond to the binary symbols 0 and 1 respectively, and n is the number of bits of the PN code.

3. A cable fault location method for reflected wave decomposition and kurtosis sorting according to claim 2, characterized in that: The frequency f of the sine wave sin The value range is [30MHz, 80MHz], and the chip duration of the PN code is T c The value range of is [0.01μs, 0.1μs], and the value range of the number of bits n of the PN code is [23,39].

4. A cable fault location method for reflected wave decomposition and kurtosis sorting according to claim 1, characterized in that: The step S2. decomposing the first reflected signal by using improved VMD to obtain K intrinsic mode function (IMF) components specifically includes: S2.1 defines the first reflection signal f(t) inside the cable, which includes the actual reflection signal and the noise signal. The formula is as follows: f(t)=αs ref (t)+n(t) In the formula, s ref (t) is the actual reflection signal, n(t) is the noise signal, and α is the weight; S2.2 introduces the Lagrangian multiplication operator λ and the quadratic penalty factor β to transform the constrained variational VMD decomposition problem into an unconstrained variational problem, and constructs the augmented form of the corresponding Lagrangian expression: Where k is the number of IMF components, u k 、w k is the kth IMF after decomposition and its corresponding central frequency, δ(t) is the Dirac function, λ is the Lagrangian factor, and L is the decomposed IMF set; S2.3 uses the center frequency comparison method to determine the number k of the optimal IMF components; S2.4 compares the kurtosis of k IMFs and determines that the one with the largest kurtosis value is defined as the second reflected signal inside the cable.

5. A cable fault location method of reflected wave decomposition and kurtosis sorting according to claim 4, characterized in that: The center frequency comparison method in step 2.4 specifically includes: S2.4.1 Determine VMD parameters, preset the number of IMFs, bandwidth constraint / penalty factor, noise tolerance, and convergence criteria; S2.4.2 performs VMD decomposition on the first reflected signal inside the cable to obtain an IMF component; S2.4.3 Use fast Fourier transform FFT to calculate the center frequency of each IMF and determine whether there is aliasing at the center frequency of each IMF. If there is no cross term at the center frequency, k is increased by 1 and returns to step S2.4.2; when there is a cross term at the center frequency, the decomposition stops, and k-1 is the optimal IMF component.

6. A cable fault location method for reflected wave decomposition and kurtosis sorting according to claim 1, characterized in that: The step S3. applying short-time Fourier transform and spectrum redistribution technology to the second reflected signal to obtain a time-frequency distribution curve of the second reflected signal specifically includes: S3.1 performs STFT on the second reflected signal, using a Gaussian window with a length of 65 steps as a window function to obtain its time-frequency distribution; S3.2 uses spectrum redistribution technology to redistribute the spectrum of the STFT time-frequency diagram to improve the time-frequency resolution and clarity of fault characteristics.

7. A cable fault location system with reflected wave decomposition and kurtosis sorting, characterized in that: Specifically including. An incident excitation module, used for transmitting an incident excitation signal to the cable and obtaining a first reflected signal inside the cable, wherein the incident excitation signal is generated by binary phase shift keying modulation of a pseudo noise (PN) code and a sine wave; A decomposition module is used to decompose the first reflection signal inside the cable by using an improved VMD algorithm to obtain K IMF components, and select the IMF component with the largest kurtosis value as the second reflection signal inside the cable; An STFT-RSP module is used to apply short-time Fourier transform and spectrum redistribution technology to the second reflected signal to obtain a time-frequency distribution curve of the second reflected signal; The fault location output module is used to select the frequency at which the maximum energy amplitude point is located in the time-frequency distribution curve of the second reflected signal, extract the energy value and time information at this frequency, construct a two-dimensional energy-time curve, and determine and output the fault position of the cable based on the maximum energy peak time difference in the two-dimensional energy-time curve and the known wave speed in the cable.

Images