Power quality disturbance detection method for distribution network based on improved ewt and cmpe

Abstract

Based on improved EWT and CMPE, the distribution network power quality disturbance detection method includes the use of improved empirical wavelet decomposition EWT to decompose the PQ disturbance signal of the active distribution network system, filter out the noise of the PQ disturbance signal, and decompose to obtain the EWT containing characteristic information Component: The EWT component containing feature information is used as the input signal of the composite multi-scale permutation entropy CMPE algorithm, and the composite multi-scale permutation entropy CMPE algorithm is used to calculate the permutation entropy of each EWT component containing feature information, and calculate each PQ disturbance signal The entropy value matrix under each modal function; use the PCA algorithm to perform dimension reduction processing on the calculated entropy value matrix, calculate the principal component, and obtain the characteristic range of various PQ disturbance signals; according to the obtained dimension reduction processing, The eigenvalue matrix is ​​used as the input of the SVM algorithm; for the active distribution network system containing distributed energy, the PQ disturbance signal identification is carried out. The method of the invention has simple steps and accurate classification, and can improve the reliability of the distribution network.

Description

technical field [0001] The invention relates to the technical field of power quality disturbance signal detection, in particular to a method for detecting power quality disturbances in distribution networks based on improved EWT and CMPE. Background technique [0002] Renewable energy has a great impact on power quality due to its uncertainty and its penetration rate is increasing year by year. In addition to the traditional power disturbance in the high-penetration active distribution network, due to the constraints of natural conditions such as environmental climate, DG output is random. , volatility and intermittency may lead to oscillation or flicker. DG output direct current such as photovoltaic array fuel cell needs to be connected to the system through power electronic devices as an interface, which will cause a lot of harmonic pollution, and due to the low inertia of DG , the system is more susceptible to various disturbances and becomes unstable; therefore, it is mo...

AI Technical Summary

Problems solved by technology

[0002] Renewable energy has a great impact on power quality due to its uncertainty and its penetration rate is increasing year by year. In addition to the traditional power disturbance in the high-penetration active distribution network, due to the constraints of natural conditions such as environmental climate, DG output is random. , volatility and intermittency may lead to oscillation or flicker. DG output direct current such as photovoltaic array fuel cell needs to be connected to the system through power electronic devices as an interface, which will cause a lot of harmonic pollution, and due to the low inertia of DG , the system is more susceptible to various disturbances and becomes unstable; therefore, it is more important to detect and classify sub-compound disturbances in high-penetration active distribution networks

[0003] There are still some problems in the existing power quality disturbance detection technology of distributed power distribution network, for example: Hilbert-Huang transform is a combination of empirical mode decomposition and Hilbert spectrum analysis, which puts the signal in the time-frequency domain However, in addition to not having a reliable theoretical basis, it also has problems such as slow calculation speed and over-envelopment; local mean decomposition is suitable for feature extraction of non-stationary power signals in power systems, and decreases according to frequency during the iterative screening process The law decomposes the signal into a series of modal functions with physical meaning, but its disadvantages of "end point effect" and "mode aliasing" greatly affect the accuracy of the results; although the variational modal decomposition decomposes the signal from the recursive screening mode Converting to a non-recursive screening variational mode solves the above-mentioned local mean decomposition problem, but at the cost of higher computational complexity and still cannot be adaptively decomposed

Images