Partial discharge signal extraction method based on AO-VMD and K-SVD learning dictionary

Through the combination method of AO-VMD and K-SVD learning dictionary, the signal processing parameters and the dictionary are optimized, and the problem of local discharge signal extraction with severe noise interference is solved, and the accuracy of high-precision signal extraction and fault analysis in complex environments is achieved. sex.

CN118964843APending Publication Date: 2024-11-15CHENGDU SHURUI INNOVATION TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410990791.0
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-23
Publication Date
2024-11-15

AI Technical Summary

Technical Problem

The prior art is difficult to accurately extract local discharge signals in complex environments, especially in cases of severe noise interference, and it is difficult to distinguish between real signals and noise, resulting in poor detection results.

Method used

Using the method based on AO-VMD and K-SVD learning dictionary, the VMD parameters are optimized through AO-VMD to perform signal noise reduction, and the learning dictionary is constructed and updated, and combined with the orthogonal matching tracking algorithm, the noise-free local discharge signal is gradually decomposed and reconstructed.

Benefits of technology

It significantly improves the accuracy and processing effect of signal extraction, and can stably extract more accurate discharge signals in high-noise environments, enhancing the accuracy of subsequent fault analysis.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118964843A_ABST
    Figure CN118964843A_ABST
Patent Text Reader

Abstract

The invention discloses a partial discharge signal extraction method based on an AO-VMD and K-SVD learning dictionary, and the method comprises the steps: S10, inputting a partial discharge signal with a noise signal, carrying out the noise reduction of an original signal through the AO-VMD, and highlighting the characteristics of the original signal; the optimal parameter of the VMD is found through the AO algorithm, and the VMD algorithm reconstructs the signal under the guidance of the optimal parameter; s20, constructing an initial learning dictionary, and updating the learning dictionary by using K-SVD; capturing original signal characteristics by using a K-SVD (K-Singular Value Decomposition) algorithm, and decomposing and reconstructing a noise-free PD signal; and S30, using an orthogonal matching pursuit algorithm to update the learning dictionary and decompose the reconstructed signal, and outputting the reconstructed partial discharge signal. According to the invention, more accurate discharge signals can be stably extracted in a more complex environment.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention belongs to the technical field of local amplification signal extraction, and in particular relates to a local discharge signal extraction method based on AO-VMD and K-SVD learning dictionaries. Background Art

[0002] Partial discharge (PD) can reflect the insulation condition of equipment and its possible failure rate. PD detection has become an important part of online monitoring of power equipment. In power systems, PD is caused by various defects in the insulation material, such as voids and electrical trees. If these phenomena are not detected and processed in time, these PD activities may lead to insulation failure of the power system, or even cause more serious problems, such as fire or equipment explosion.

[0003] However, it is not easy to detect and extract PD signals: in the early insulation failure stage of the equipment, the PD signal is usually very weak, and there may be a lot of electromagnetic noise at the detection site. As a result, the detected PD signal is usually interfered by a lot of noise, which seriously affects the effect of partial discharge monitoring and the determination of insulation conditions. At the same time, with the increasing complexity of the current power system, the interference of PD signals is also increasing, so it is necessary to develop more efficient methods to distinguish between true partial discharge signals and noise generated by other sources.

[0004] The detection of PD signals faces interference from many complex factors, among which noise pollution is particularly serious, mainly reflected in white Gaussian noise and narrowband noise. Therefore, researchers are particularly concerned about how to eliminate noise, determine edges, and accurately extract the waveform of PD signals. In recent years, a series of progress has been made in the processing technology of PD signals. At present, various types of methods have been proposed to reduce white noise and narrowband noise in PD signals, including time domain processing methods, time-frequency domain analysis methods, mathematical analysis methods, and machine learning algorithms.

[0005] As a basic technology in the time domain analysis of PD signals, filtering technology can extract valuable information and eliminate unnecessary noise to a certain extent. However, the useful signal and noise in the noisy PD signal have certain similarities in the time domain and are difficult to distinguish. Therefore, when using filters to process signals, it is very likely that part of the useful signal will be mistakenly filtered out. Once the useful signal is lost, it will have an adverse effect on the fault analysis and judgment after the useful signal is extracted.

[0006] Wavelet method and EMD method are both very representative time-frequency domain analysis methods. If the mother wavelet is not selected properly, it may affect the extraction accuracy of PD signal. EMD method cannot completely eliminate the influence of modal aliasing.

[0007] Mathematical analysis based on the SVD concept has been used many times in the processing of PD signals. The advantage of the SVD algorithm is that it can effectively extract useful features from complex data and reduce the dimension of the data without losing too much information. However, when the characteristics of the PD signal are not obvious and difficult to distinguish from noise, SVD may not be able to distinguish which is noise and which is a valuable signal. At the same time, the SVD algorithm principle is based on the mathematical principle of linear algebra, but the generation and propagation of PD signals may involve nonlinear processes, thereby generating nonlinear data. For nonlinear data, the processing effect of SVD is not good.

[0008] Machine learning algorithms have been continuously proposed and applied, providing a new idea for the field of signal processing. However, in PD signal processing, there are few training samples available and the data is variable. In this case, it may be difficult for machine learning algorithms to obtain good training results. Summary of the invention

[0009] In order to solve the above problems, the present invention proposes a local discharge signal extraction method based on AO-VMD and K-SVD learning dictionary, which can stably extract more accurate discharge signals in a more complex environment.

[0010] To achieve the above object, the technical solution adopted by the present invention is: a method for extracting partial discharge signals based on AO-VMD and K-SVD learning dictionary, comprising the steps of:

[0011] S10, inputting a local discharge signal with noise signals, firstly using AO-VMD to reduce the noise of the original signal, highlighting the characteristics of the original signal; finding the optimal parameters of VMD through the AO algorithm, and the VMD algorithm reconstructing the signal under the guidance of the optimal parameters;

[0012] S20, construct an initial learning dictionary and use K-SVD to update the learning dictionary; use the K-SVD algorithm to capture the original signal characteristics and decompose and reconstruct the noise-free PD signal;

[0013] S30, using the orthogonal matching pursuit algorithm to update the learning dictionary and decompose the reconstructed signal, and output the reconstructed partial discharge signal.

[0014] Furthermore, the AO algorithm processing process includes the steps of:

[0015] S101, AO parameter setting and VMD parameter setting;

[0016] S102, using AO to optimize VMD, and obtaining VMD optimization parameters through iterative calculation.

[0017] Furthermore, AO is used to optimize VMD to obtain VMD optimization parameters, including the steps of:

[0018] Determine whether the maximum number of iterations has been reached, otherwise proceed to the next step;

[0019] Determine whether the fitness value becomes smaller, if so, update the VMD parameters and increase the number of iterations by 1, otherwise keep the VMD parameters and increase the number of iterations by 1;

[0020] The above steps are repeated until the maximum number of iterations is reached, at which point the VMD optimization parameters are output as the VMD optimal parameters.

[0021] Furthermore, reconstructing the signal using VMD includes the steps of: constructing a variational problem model to achieve simultaneous decomposition of the signal into multiple finite bandwidth modes, each mode is concentrated around a corresponding center frequency; decomposing the contaminated PD signal into multiple modes through a VMD algorithm; and extracting a noise-free local discharge signal.

[0022] Furthermore, updating the learning dictionary using K-SVD includes the following steps:

[0023] S21, dictionary initialization: initialized by generating a random matrix and then performing column normalization to form the initial dictionary;

[0024] S22, sparse coding: sparse coding is performed on the initial dictionary through the MP greedy algorithm;

[0025] S23, dictionary update: optimize the representation capability of dictionary atoms through iteration to adapt to the characteristics of PD signals.

[0026] Furthermore, the orthogonal matching pursuit algorithm adopts an adaptive threshold segmented OMP algorithm, which removes the fixed number of iterations of the original OMP algorithm and adds a continuous iteration loop until the norm of the residual is lower than the set threshold.

[0027] The beneficial effects of adopting this technical solution are:

[0028] In addition to updating the dictionary, the algorithm of the present invention also updates the representation of the signal to improve the processing accuracy. At the same time, this algorithm is also iterative, and each step is aimed at reducing the reconstruction error until the algorithm converges and outputs the final processing result, which can stably extract more accurate discharge signals in more complex environments.

[0029] The AO-VMD algorithm proposed in the present invention is a VMD optimization algorithm that improves the preprocessing effect by optimizing the parameters of VMD. By decomposing and reconstructing the PD signal obliterated by noise through AO-VMD, its key features are more easily captured by the K-SVD algorithm, thereby enhancing the performance and accuracy of subsequent processing steps. This combination method can significantly improve the processing effect when processing complex signals, especially signals in high-noise environments. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 A schematic flow chart of a partial discharge signal extraction method based on AO-VMD and K-SVD learning dictionary of the present invention;

[0031] Figure 2 Schematic diagram of the process of the AO-VMD algorithm in an embodiment of the present invention. DETAILED DESCRIPTION

[0032] In order to make the purpose, technical solutions and advantages of the present invention more clear, the present invention is further described below with reference to the accompanying drawings.

[0033] In this embodiment, see Figure 1 As shown, the present invention proposes a method for extracting partial discharge signals based on AO-VMD and K-SVD learning dictionaries, comprising the steps of:

[0034] S10, inputting a local discharge signal with noise signals, firstly using AO-VMD to reduce the noise of the original signal, highlighting the characteristics of the original signal; finding the optimal parameters of VMD through the AO algorithm, and the VMD algorithm reconstructing the signal under the guidance of the optimal parameters;

[0035] S20, construct an initial learning dictionary and use K-SVD to update the learning dictionary; use the K-SVD algorithm to capture the original signal characteristics and decompose and reconstruct the noise-free PD signal;

[0036] S30, using the orthogonal matching pursuit algorithm to update the learning dictionary and decompose the reconstructed signal, and output the reconstructed partial discharge signal.

[0037] As the optimization method of the above embodiment, the AO algorithm processing process includes the steps of:

[0038] S101, AO parameter setting and VMD parameter setting;

[0039] S102, using AO to optimize VMD, and obtaining VMD optimization parameters through iterative calculation.

[0040] Among them, Figure 2 As shown, when using AO to optimize VMD, the VMD optimization parameters are obtained, including the steps:

[0041] Determine whether the maximum number of iterations has been reached, otherwise proceed to the next step;

[0042] Determine whether the fitness value becomes smaller, if so, update the VMD parameters and increase the number of iterations by 1, otherwise keep the VMD parameters and increase the number of iterations by 1;

[0043] The above steps are repeated until the maximum number of iterations is reached, at which point the VMD optimization parameters are output as the VMD optimal parameters.

[0044] Among them, the signal reconstruction using VMD includes the following steps: constructing a variational problem model to realize the simultaneous decomposition of the signal into multiple finite bandwidth modes, each mode is concentrated around the corresponding center frequency; decomposing the contaminated PD signal into multiple modes through the VMD algorithm; and extracting the noise-free local discharge signal.

[0045] AO-VMD is an advanced signal processing method that combines orthogonal array optimization methods and variational mode decomposition techniques. The algorithm aims to improve the quality and accuracy of signal decomposition by optimizing the parameters of VMD. In this method, the orthogonal array is used to effectively explore and evaluate the parameter space to find the most optimized VMD parameter configuration. This configuration includes determining the most appropriate decomposition level and penalty factor, both of which are crucial to the decomposition quality of the signal. Through this optimization, AO-VMD can more accurately extract key features from complex signals and improve the effect and efficiency of signal processing. AO-VMD is a very effective signal processing tool in high noise or non-stationary signal environments.

[0046] As an optimization method of the above embodiment, using K-SVD to update the learning dictionary includes the following steps:

[0047] S21, dictionary initialization: initialized by generating a random matrix and then performing column normalization to form the initial dictionary;

[0048] S22, sparse coding: sparse coding is performed on the initial dictionary through the MP greedy algorithm;

[0049] S23, dictionary update: optimize the representation capability of dictionary atoms through iteration to adapt to the characteristics of PD signals.

[0050] K-SVD is an iterative algorithm that gradually improves the representation ability of the dictionary by alternately performing sparse coding and dictionary updates. Its adaptive nature allows the basic dictionary to capture more details from the signal, better grasp the core features of the PD signal, and continuously update itself. In addition, K-SVD also supports regularization, which helps control the overfitting of the dictionary and ensure stable and accurate signal extraction.

[0051] The present invention generates a matrix filled with random numbers as an initial dictionary. Because the matrix is ​​drawn from a standard normal distribution, the values ​​in the matrix will be concentrated around 0, but there will also be some larger or smaller values. The random matrix columns are then normalized to ensure that the length (or "energy") of each column is 1. Such an initialization method is the most commonly used initial dictionary generation method in K-SVD, and it does not utilize any prior knowledge about the original signal. At the same time, this random initialization method is unsupervised, which means that the method does not rely on any specific attributes of the training data.

[0052] As an optimization method of the above embodiment, the orthogonal matching pursuit algorithm adopts an adaptive threshold segmented OMP algorithm, which removes the fixed number of iterations of the original OMP algorithm and adds a continuous iteration loop until the norm of the residual is lower than the set threshold.

[0053] The improved OMP algorithm of the present invention is more flexible and can dynamically determine the number of iterations based on data, so that the dictionary is more complete and the characteristics of the PD signal are better restored.

[0054] The above shows and describes the basic principles and main features of the present invention and the advantages of the present invention. It should be understood by those skilled in the art that the present invention is not limited to the above embodiments. The above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention to be protected. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims

1. A partial discharge signal extraction method based on AO-VMD and K-SVD learning dictionary, It is characterized in that Includes steps: S10, inputting a partial discharge signal with noise signals, firstly using AO-VMD to reduce the noise of the original signal, highlighting the characteristics of the original signal; finding the optimal parameters of VMD through the AO algorithm, and the VMD algorithm reconstructing the signal under the guidance of the optimal parameters; S20, construct an initial learning dictionary and use K-SVD to update the learning dictionary; The K-SVD algorithm is used to capture the original signal characteristics and decompose and reconstruct the noise-free PD signal; S30, using the orthogonal matching pursuit algorithm to update the learning dictionary and decompose the reconstructed signal, and output the reconstructed partial discharge signal.

2. A method for extracting partial discharge signals based on AO-VMD and K-SVD learning dictionaries according to claim 1, It is characterized in that The AO algorithm processing process includes the following steps: S101, AO parameter setting and VMD parameter setting; S102, using AO to optimize VMD, and obtaining VMD optimization parameters through iterative calculation.

3. A method for extracting partial discharge signals based on AO-VMD and K-SVD learning dictionary according to claim 2, It is characterized in that When using AO to optimize VMD, the VMD optimization parameters are obtained, including the following steps: Determine whether the maximum number of iterations has been reached, otherwise proceed to the next step; Determine whether the fitness value becomes smaller, if so, update the VMD parameters and increase the number of iterations by 1, otherwise keep the VMD parameters and increase the number of iterations by 1; The above steps are repeated until the maximum number of iterations is reached, at which point the VMD optimization parameters are output as the VMD optimal parameters.

4. The method for extracting partial discharge signals based on AO-VMD and K-SVD learning dictionary according to claim 1, It is characterized in that Reconstructing the signal using VMD includes the following steps: constructing a variational problem model to achieve simultaneous decomposition of the signal into multiple finite bandwidth modes, each mode is concentrated around the corresponding center frequency; decomposing the contaminated PD signal into multiple modes through the VMD algorithm; and extracting the noise-free local discharge signal.

5. The method for extracting partial discharge signals based on AO-VMD and K-SVD learning dictionary according to claim 1, It is characterized in that Updating the learning dictionary using K-SVD includes the following steps: S21, dictionary initialization: initialized by generating a random matrix and then performing column normalization to form the initial dictionary; S22, sparse coding: sparse coding is performed on the initial dictionary through the MP greedy algorithm; S23, dictionary update: optimize the representation capability of dictionary atoms through iteration to adapt to the characteristics of PD signals.

6. A method for extracting partial discharge signals based on AO-VMD and K-SVD learning dictionary according to claim 1, It is characterized in that The orthogonal matching pursuit algorithm adopts an adaptive threshold segmented OMP algorithm, removes the fixed number of iterations of the original OMP algorithm, and adds a continuous iteration cycle until the norm of the residual is lower than the set threshold.