PD feature extraction based on cross-wavelet transform and PCA

Abstract

Aimed at the high sensitivity to noise in partial discharge (PD) pattern recognition, a novel feature extraction method based on cross-wavelet transform and Principal Component Analysis is proposed. Cross-wavelet transform of PD signals provides new characteristic parameters for pattern recognition. Principal Component Analysis (PCA) is used to reduce dimensions of the new selected parameters. Back Propagation Neural Network, Radial Basis Function Neural Network and Probabilistic Neural Network classifiers are utilized for PD pattern recognition. Results demonstrate that, the proposed feature extraction approach can reduce the influence of noises. And the recognition results are encouraging with a reasonable accuracy.