CN-121997021-A - Low-voltage power supply cable fault feature extraction method and system

CN121997021ACN 121997021 ACN121997021 ACN 121997021ACN-121997021-A

Abstract

The invention discloses a low-voltage power supply cable fault feature extraction method and a system. The method comprises the steps of obtaining fault current signals of a low-voltage power supply cable, determining the optimal mode number by taking the minimum mutual information as a principle, optimizing parameters of variation mode decomposition by adopting an improved black-wing iris optimization algorithm, optimizing penalty coefficients by taking the minimum envelope entropy as a fitness function, combining the optimal mode number and the optimized penalty coefficients to form optimal VMD parameter combination, performing VMD decomposition on the fault current signals based on the optimal VMD parameter combination to obtain a plurality of inherent mode components, performing Hilbert transform on the inherent mode components, and extracting amplitude and phase change characteristics to obtain fault characteristics. By implementing the method, the parameters of the VMD are more intelligently and accurately optimized, so that the accuracy and reliability of low-voltage power supply cable fault feature extraction are improved.

Inventors

CAO JINGXING
ZHOU LI
TAN XIAO
LI CHENYING
ZHANG WEI
WANG QI
WU SHUQUN
ZHANG YIMING

Assignees

国网江苏省电力有限公司电力科学研究院
国网江苏省电力有限公司
江苏省电力试验研究院有限公司
东南大学溧阳研究院

Dates

Publication Date: 20260508
Application Date: 20260121

Claims (10)

1. The low-voltage power supply cable fault feature extraction method is characterized by comprising the following steps of: Acquiring a fault current signal of a low-voltage power supply cable; Determining the number of optimal modes by taking the minimum mutual information as a principle; Optimizing parameters of variation modal decomposition by adopting an improved black-fin iris optimization algorithm, and optimizing a penalty coefficient by taking minimum envelope entropy as an fitness function; combining the optimal modal number and the optimized penalty coefficients to form an optimal VMD parameter combination; performing VMD decomposition on the fault current signal based on the optimal VMD parameter combination to obtain a plurality of inherent modal components; and performing Hilbert transformation on the inherent modal component, and extracting amplitude and phase change characteristics to obtain fault characteristics.
2. The method for extracting fault characteristics of a low-voltage power supply cable according to claim 1, wherein determining the number of optimal modes based on the principle of minimum mutual information comprises: Constructing an evaluation function for determining the optimal modal decomposition layer number based on the concept of mutual information; And calculating the modal decomposition layer number which enables the evaluation function to reach the minimum value so as to obtain the optimal modal number.
3. The method for extracting fault characteristics of a low-voltage power supply cable according to claim 2, wherein the evaluation function is used for calculating a ratio between a cumulative sum of mutual information of each mode and a cumulative sum of mutual information between an original current signal and each decomposition mode, and an expression of the evaluation function is: wherein, the method comprises the steps of, As an input signal to the VMD, Indicating that the VMD decomposition is performed to obtain the first An IMF component of the number; indicating that the VMD decomposition is performed to obtain the first IMF component of individual Representing the sum of mutual information between the different modal components, Representing the sum of the mutual information between all modal components and the original signal.
4. The method for extracting fault characteristics of a low-voltage power supply cable according to claim 1, wherein optimizing parameters of variation modal decomposition by using an improved black-fin iris optimization algorithm and optimizing penalty coefficients by using minimum envelope entropy as an fitness function comprises: optimizing the punishment coefficient in the global search range by adopting an improved black-fin iris optimization algorithm, and selecting envelope entropy as a fitness function to determine the optimized punishment coefficient.
5. The method for extracting fault characteristics of a low voltage power supply cable according to claim 4, wherein optimizing the penalty coefficient in a global search range by using an improved black-fin iris optimization algorithm, and selecting an envelope entropy as a fitness function to determine the optimized penalty coefficient comprises: randomly generating a group of iris nigromaculata individuals representing punishment coefficient values, and determining the individual with the highest fitness as a leader; Searching a punishment coefficient value area with the lowest envelope entropy value by updating the individual position to simulate the process of approaching the iris nigromaculata to the prey; adjusting a leader according to individual fitness and introducing cauchy mutations to increase population diversity; Using VMD decomposition signals for each penalty coefficient value, and calculating the envelope entropy of the obtained IMF component to evaluate the fitness of each penalty coefficient; and selecting the penalty coefficient value which generates the lowest envelope entropy value as the global optimal penalty coefficient.
6. The method for extracting fault characteristics of low voltage power supply cable according to claim 5, wherein the calculation formula of the envelope entropy of the obtained IMF component is , wherein, Representing an envelope entropy value of the signal; Represent the first Envelope signals obtained after the demodulation operation of the IMF components; Is that Is a normalization operation of (a).
7. The method for extracting fault characteristics of a low-voltage power supply cable according to claim 1, wherein the VMD-decomposing the fault current signal based on the optimal VMD parameter combination to obtain a plurality of natural modal components includes: And taking the fault current signal as input data of a VMD algorithm, and performing variational mode decomposition operation on the input data by utilizing the optimal VMD parameter combination to obtain a plurality of natural mode components with respective center frequencies and limited bandwidths, wherein each natural mode component represents an independent characteristic frequency band in an original signal.
8. The method for extracting fault characteristics of a low-voltage power supply cable according to claim 1, wherein the performing hilbert transformation on the natural modal component to extract the amplitude and phase variation characteristics to obtain the fault characteristics includes: Re-selecting a specific modal component containing the injection characteristic frequency from the natural modal component; Performing Hilbert transform on the specific modal component and the corresponding power supply voltage signal to generate an analytic signal, wherein the analytic signal comprises amplitude change information; calculating the instantaneous phase of the analytic signal and the fault current signal; Determining a phase difference between a supply voltage and a current using the instantaneous phase; converting the phase difference from radian to angle system to obtain phase difference information; And identifying and extracting fault characteristics according to the amplitude change information and the phase difference information.
9. The method for extracting fault characteristics of low voltage power supply cable according to claim 8, wherein the instantaneous phase of the voltage is expressed as , wherein, The corresponding analysis signal is the supply voltage of the injection signal ; To inject the fault current of the signal, the corresponding analysis signal is Instantaneous phase of current The method comprises the following steps: the phase difference is as follows: according to Will be And converting the phase difference information into an angle system unit to specifically characterize the phase difference information of the fault signal.
10. The utility model provides a low voltage power supply cable fault feature draws system which characterized in that includes: the acquisition unit is used for acquiring fault current signals of the low-voltage power supply cable; The number determining unit is used for determining the number of the optimal modes by taking the minimum mutual information as a principle; the coefficient determining unit is used for optimizing parameters of variation modal decomposition by adopting an improved iris blackwing optimization algorithm, and optimizing a penalty coefficient by taking minimum envelope entropy as an fitness function; The parameter combination determining unit is used for combining the optimal modal number and the optimized penalty coefficient to form an optimal VMD parameter combination; The decomposition unit is used for performing VMD decomposition on the fault current signal based on the optimal VMD parameter combination to obtain a plurality of inherent modal components; And the extraction unit is used for carrying out Hilbert transformation on the inherent modal component and extracting amplitude and phase change characteristics so as to obtain fault characteristics.

Description

Low-voltage power supply cable fault feature extraction method and system Technical Field The invention relates to the technical field of fault diagnosis of low-voltage cables, in particular to a low-voltage power supply cable fault feature extraction method and system. Background The stability and reliability of the low-voltage power supply cable as a key component of the urban power distribution network are directly related to the energy safety of the city and the life quality of residents. These cables are prone to potential insulation defects that can occur when exposed to complex environmental conditions for long periods of time, thereby causing early failure. To prevent such problems, prior art solutions generally rely on injecting characteristic signals of a specific frequency into the cable and identifying possible fault points by monitoring the variation of these signals. However, this approach presents challenges during actual operation. Firstly, the characteristic signals and the power frequency signals in the injection cable are mutually overlapped, so that the complexity of signal processing is increased. Second, it becomes more difficult to accurately extract useful fault signature due to external interference and noise. In this case, how to effectively separate the injected characteristic signal from the mixed signal becomes an important link for improving the fault diagnosis accuracy. The variable modal decomposition (VMD, variational Mode Decomposition) as an advanced adaptive signal processing method can decompose the composite signal into a plurality of independent narrowband intrinsic mode components, which provides a possibility to solve the above-mentioned problems. The effectiveness of the VMD algorithm is highly dependent on the choice of two key parameters, the number of modes K and the penalty factor α. If these two parameters are not properly selected, the problem of under-decomposition or over-decomposition may result, i.e., either not all signal components can be completely separated or redundant components may be generated to affect the signal analysis results. Traditionally, these two parameters are often set empirically, which limits the effectiveness of the VMD in practical applications. Therefore, it is necessary to design a new method, which is more intelligent and more accurate to optimize the parameters of the VMD, so as to improve the accuracy and reliability of the fault feature extraction of the low-voltage power supply cable. Disclosure of Invention The invention aims to overcome the defects of the prior art and provides a low-voltage power supply cable fault feature extraction method and system. In order to achieve the purpose, the invention adopts the following technical scheme that the low-voltage power supply cable fault feature extraction method comprises the following steps: Acquiring a fault current signal of a low-voltage power supply cable; Determining the number of optimal modes by taking the minimum mutual information as a principle; Optimizing parameters of variation modal decomposition by adopting an improved black-fin iris optimization algorithm, and optimizing a penalty coefficient by taking minimum envelope entropy as an fitness function; combining the optimal modal number and the optimized penalty coefficients to form an optimal VMD parameter combination; performing VMD decomposition on the fault current signal based on the optimal VMD parameter combination to obtain a plurality of inherent modal components; and performing Hilbert transformation on the inherent modal component, and extracting amplitude and phase change characteristics to obtain fault characteristics. The method for determining the optimal mode number based on the principle of minimum mutual information comprises the following steps: Constructing an evaluation function for determining the optimal modal decomposition layer number based on the concept of mutual information; And calculating the modal decomposition layer number which enables the evaluation function to reach the minimum value so as to obtain the optimal modal number. The evaluation function is used for calculating the ratio between the accumulated sum of the mutual information of each mode and the accumulated sum of the mutual information between the original current signal and each decomposition mode, wherein the expression of the evaluation function is as follows: wherein, the method comprises the steps of, As an input signal to the VMD,Indicating that the VMD decomposition is performed to obtain the firstIMF components of the individual.Indicating that the VMD decomposition is performed to obtain the firstIMF component of individualRepresenting the sum of mutual information between the different modal components,Representing the sum of the mutual information between all modal components and the original signal. The further technical scheme is that the optimizing parameters of the variational modal decomposition by adopt