A Fault Classification and Direction Estimation Algorithm by Neural Network

신경회로망을 이용한 송전선로 보호용 방향 개전 및 고장상 선택 알고리즘

The direction and the type of a fault on a transmission line needs to be identified rapidly and correctly. This paper presents a approach to identify fault direction and type with neural network on double circuit transmission line. A neural network based on self organization map(SOM) provides the ability to accurately classify the fault type and to select of a fault direction. In this paper, proposed algorithm uses different patterns of the associated voltages and currents in order to identify fault clusters.