A decision tree-based method for power system fault diagnosis by synchronized Phasor Measurements

Abstract

This paper introduces a novel approach for power system fault diagnosis based on synchronized phasor measurements during the fault. The synchronized measurements are obtained in real time from Phasor Measurement Units (PMUs) and compared with offline thresholds determined by decision trees (DTs) to diagnose the fault. The DTs have already been trained offline using detailed power system analysis for different fault cases. While the traditional methods for fault diagnosis use the status of protective relays (PRs) and circuit breakers (CBs) to infer the fault section in the power system, the proposed method uses the available signals following the fault and thus can be trusted even in case of protection system failure. Therefore, fault diagnosis can be achieved successfully especially in blackout situations, where many CBs and and/or PRs malfunction. Simulation results on the IEEE 118-bus test system show that the proposed method is highly accurate in wide area fault diagnosis in spite of incomplete system observability.