PRPD data analysis with Auto-Encoder Network

Abstract

Gas Insulated Switchgear (GIS) is related to the stable operation of power equipment. The traditional partial discharge pattern recognition method relies on expert experience to carry out feature engineering design artificial features, which has strong subjectivity and large blindness. To address the problem, we introduce an encoding-decoding network to reconstruct the input data and then treat the encoded network output as a partial discharge signal feature. The adaptive feature mining ability of the Auto-Encoder Network is effectively utilized, and the traditional classifier is connected to realize the effective combination of the deep learning method and the traditional machine learning method. The results show that the features extracted based on this method have better recognition than artificial features, which can effectively improve the recognition accuracy of partial discharge.

A MobileNets Convolutional Neural Network for GIS Partial Discharge Pattern Recognition in the Ubiquitous Power Internet of Things Context: Optimization, Comparison, and Application

“…The existing feature construction methods mainly include Fourier transforms, wavelet transforms, empirical mode decomposition, S-parameter transformation, and polar coordinate transformation [10]- [13]. Furthermore, in order to effectively extract the most critical characteristic parameters of the PD and reduce the feature dimension, principal component analysis (PCA) and an auto-encoder have been introduced for GIS PD recognition and classification [14], [15].…”