S-transform Based Support Vector Regression for Detection of Incipient Faults and Voltage Disturbances in Power Distribution Networks

摘要

Many of the electrical systems throughout the world are experiencing problems with aging insulation. When an insulation failure occurs, it can cause sustained interruption which leads to production loss, expensive equipment damage and substantial financial loses. With the ability to identify incipient fault by detecting potential insulation failures before they occur, power utility's engineer will be able to reduce customers lost profit opportunities. In this paper, a new technique to detect the occurrence of incipient fault and voltage disturbances is proposed. The technique uses the S-transform and the Support Vector Regression (SVR) to extract features from the recorded voltage and currents waveforms and to detect the potential occurrences of incipient fault and voltage disturbance. A case study is presented to evaluate the accuracy of the S-transform based SVR in detecting incipient faults and voltage disturbances occurring in the power distribution networks.