APPLICATION OF DSP FOR DISTRIBUTION SYSTEM PROTECTION

Abstract

Fault occurs frequently on Electrical Distribution system. The main objective of the work is to develop an efficient protection algorithm for Distribution system protection that includes fault detection, classification and location. The total protection scheme consists of two modules. First module is for the detection and classification of faults; and the second module is for estimating the faulted section and location of faults. Wavelet Packet Transform coefficients are used for the features extraction for detection and classification of faults module. The obtained features are fed to Artificial Neural Network (ANN) for fault classification. Entropy in the wavelet packet energy spectrum are used for the feature extraction and are given to Artificial Neural Network (ANN) fault section estimator and fault locator for estimating the fault location. For testing the proposed method 25kV SaskPower distribution system is considered. The considered model has been simulated using PSCAD software.