PCA-LDA for partial discharge classification on high voltage equipment

Abstract

This document proposes a statistical classification model using principal component analysis (PCA) for a data reduction approach combined with linear discriminant analysis (LDA) for a classification purpose, so called, PCA-LDA to classify PD patterns into four categories listed as corona at high voltage side in air, corona at low voltage side in air, surface in air, and internal discharge. The original independent variables in this classification model are skewness, kurtosis, asymmetry, and cross correlation following the φ-q-n PD patterns obtained from the fingerprint analysis which is a digital signal processing technique for PD measurement. The experiments were set to simulate all four PD patterns to obtain statistical parameters into 10 independent variables from the fingerprint analysis. With PCA, the original 10 independent variables can be reduced to 3 independent variables and using PCA can retains the accuracy of this classification model to 97.5 percent. © 2006 IEEE.