Classification and Analysis of Power System Faults in IEEE-14 Bus System using Machine learning Algorithm

Abstract

The complexity in the Power transmission and distribution sectors is increasing day by day with a continuous increase in power demand. Due to the increased complexity, the frequency of symmetrical and unsymmetrical fault has significantly increased, leading to frequent tripping of circuit Breakers. Due to which the reliability and quality provided at the consumer end are jeopardizing, therefore, it is necessary to develop an intelligent mechanism that can easily identify the power system fault so that occurrence of failure is more frequent in transmission and distribution lines due to symmetrical and unsymmetrical faults. As a result, the power reliability and the quality of the service provided by the Power Sector are identified to be at risk. Providing an adequate system that can identify the power system faults is very essential for the faster restoration of the faults in the power supply. In this paper, the analysis and performance of machine learning algorithms for the classification of faults in the power system are studied. Current and Voltage data were taken from a standard IEEE-14 bus system normal, symmetrical, and unsymmetrical fault cases were simulated in MATLAB Simulink. The data extracted is then used to train and tested through the SVM model. The main objective of this work is to classify the different types of fault happening in power systems with accuracy for faster restoration.