Detection and Classification of Power Quality Events Using Wavelet Energy Difference and Support Vector Machine

Abstract

Detection and classification of power quality events (PQE) to improve the quality of electric power is an important issue in utilities and industrial factories. In this paper, an approach to classify PQE with noise based on wavelet energy difference and support vector machine (SVM) is presented. Here PQE signals are decomposed into ten layers by db4 wavelet with multi-resolution. Energy differences (ED) of every level between PQE signal and standard signal are extracted as eigenvectors. Principal component analysis (PCA) is adopted to reduce the dimensions of eigenvectors and find out the main structure of the matrix, which forms new feature vectors. Then these new feature vectors are divided into two groups, namely training set and testing set. The method of cross-validation is used for the training set to select the optimal parameters adaptively and construct the training model. Also, the testing set is substituted into the training model for testing. Finally, the proposed method results are compared with S-transform (ST)- and Hilbert-Huang transform (HHT)-based PQE classification to verify the accuracy of classification. The results demonstrated show that the proposed method has high accuracy, strong resistance to noise, and fast classification speed and is suitable for the detection and classification of PQE.