IDENTIFICATION AND CLASSIFICATION OF DISTURBANCES ON HV TRANSMISSION SYSTEMS

Published in 2015

This publication is free for authenticated users

Summary

Availability of electrical power has much more importance than before and there is no tolerance to power interrupts in today's power networks. There are many different types of transient disturbances and short duration variations affecting the safe operation of sub-components such as high voltage transmission lines, underground power cables, switching equipment, etc. It is necessary to identify and classify those in a very short time in order to provide high speed clearance of the disturbance by taking necessary action such as isolating the faulty line section to enable fast restoration. Therefore, it is needed to develop an algorithm and analyzing scheme that can be used on-line monitoring serving for identification of transient events and their classification by defining the possible underlying root cause. In this paper, a new scheme has been developed for identification and classification of disturbances on transmission lines by utilizing a scheme that is based on wavelet based sub-band energy distribution for feature extraction and artificial neural network (ANN) for classification of disturbances. After defining wavelet related parameters which reveal better classification scheme, energy distribution of wavelet coefficients at each decomposition level was determined and used as transient feature identifier presented to a neural network classifier. Different training algorithms have been tried in ANN and useful ones for this purpose have been identified. Simulation studies have been carried out in order to confirm the effectiveness of the proposed scheme by using common disturbance types on a transmission line. The results indicate that the proposed scheme is effective and able to identify and classify common disturbances with enough accuracy and can be adopted as a line protection algorithm.