A Power Swing Detection Scheme in Distance Relays Using Support Vector Machine Based Classifier

Authors
1 mashhad university
2 gonabad
Abstract
In power systems, distance relays are utilized widely as main protection of the high voltage transmission lines, and their safe operation has significant effects on health of the components and stability of the system. Power swing is a phenomenon that can cause mal-operation of the distance relays while there is no fault in the system. In this paper, a protection scheme based on Support Vector Machine (SVM) classifier is proposed for the digital distance relays to distinguish between power swings and faults. In this work, simulations are done on a study system and related train and test patterns are generated to analyze performance of the proposed method. Also, the proper structure of the SVM classifier is investigated using the train and test patterns. The train and test patterns contain information of the different power swing conditions and symmetrical and asymmetrical types of faults. The results of the tests and simulations confirm the efficiency of the proposed protection algorithm

Keywords