Electric Transmission System Fault Identification Using Artificial Neural Networks

Abstract

Electric transmission systems are complex mesh networks that direct large amounts of energy from the point of generation to the point of consumption. Electric faults can cripple a system as power flows must be directed around the fault therefore leading to numerous potential issues such as overloading, customer service interruptions, or cascading failures. Therefore, identifying the classification and location of these faults as quickly and efficiently as possible is crucial. This work aims to utilize artificial neural networks to determine fault type and location based on measured voltages and currents. Eventually, once developed, this solution could be utilized for fault detection and classification on several transmission circuit topologies as well as with different fault types and resistances.