AN IDENTIFICATION METHOD FOR VOLTAGE SAG IN DISTRIBUTION SYSTEMS USING SVM WITH GREY WOLF ALGORITHM, 1-10.

Wu Jiekang and Chen Xiaohua

Keywords

Distribution systems; Voltage sag source identification; Gray Wolf algorithm; Support vector machine; Empirical mode decomposition

Abstract

For the voltage sag caused by short-circuit fault, transformer switching and induction motor starting in distribution network, an optimised support vector machine (SVM) method based on Grey Wolf algorithm (GWO) is proposed for voltage sag identification. The empirical mode decomposition method is used to analyse the voltage sag signal, obtain an inherent mode function set (IMFs), and calculate the energy entropy of each order IMF as the eigenvector. In order to solve the problem that the traditional SVM is easy to fall into local optimisation in the process of optimisation, a method of optimising the penalty factor and kernel function parameters of SVM through GWO is proposed, a GWO-SVM classifier is constructed, and then the extracted feature vector is input into the GWO-SVM classifier to train and recognise the samples, so as to realise the automatic classification and identification of different types of voltage sag sources. The simulation results show the effectiveness of the extracted feature vector and GWO-SVM classifier. Compared with other five traditional methods, it is verified that it has fast speed and high precision.

Important Links:

Go Back