FAULT IDENTIFICATION IN POWER SYSTEM WITH DWT AND MULTICLASS SUPPORT VECTOR MACHINE

Powers system is now rapidly growing in size and complexity due to increase in demands. All the sectors such as power generation, power transmission, and distribution and load systems are affected due to this growth. Reliability of the electrical system is affected due to different types of faults like short circuit condition in power system network which results in economic losses and shrinks the reliable operation of power system. Electrical fault generated due to many abnormal condition such as failures of transformers and rotating machines, human errors and environmental conditions. The fault may be LG, LLG, LLL or LLLG types. In this paper, Fault detection and identification in electric power transmission has described using wavelet transform and multi-class SVM classifier. The developed method of fault identification has shown a capability of detecting a fault either it is due to single line to ground and double line to ground or all the three phases fault. The results obtained in simulation shows the validation of the proposed methodology for the scenarios considered.