Wavelet multiresolution analysis for monitoring the occurrence of arcing on overhead electrified railways

Abstract

A crucial problem for railway companies is the detection of the critical points of the overhead line for predicting excessive wear of the contact wire and for helping maintenance activities. An innovative tool, based on wavelet multiresolution analysis, is proposed in this paper for on-line monitoring of the quality of current collection. Wavelet decomposition of data collected from different measurement systems (e.g. total traction line current or signals acquired from unconventional sensors) can perform an automatic analysis of the data collected and can detect a singular occurrence of arcing due to loss of contact of the pantograph. A relevant outcome of such an analysis is the evaluation of a quantitative index for checking the quality of the current collection via an integral analysis of the multiresolution signal decomposition. Examples of analysis from experimental data are reported in the paper.