Integrated Transmission Line Fault Detection, Classification, and Relay Coordination Using Discrete Wavelet Transform

Abstract

Reliable transmission protection requires faults to be identified rapidly while preserving selectivity between primary and backup relays. This methodology paper develops an integrated framework for fault detection, fault-type identification, distance-zone decision, and relay coordination in a high-voltage transmission network. A 220-kV, 50-Hz, 100-km two-source line is used as the benchmark. Local three-phase voltage and current signals are filtered and sampled, and discrete wavelet transform (DWT) energy together with symmetrical-component indices are used to detect disturbances and distinguish line-to-ground, line-to-line, double-line-to-ground, and three-phase faults. Apparent loop impedance is then used to assign distance-protection zones, while backup directional overcurrent operation is constrained by a coordination time interval of at least 0.25 s. The approach is consistent with pre-2018 work on wavelet, artificial-intelligence, and optimized relay coordination [1]-[12]. For methodological validation, a reproducible synthetic benchmark of 96 fault cases was evaluated across fault position, resistance, and inception angle. The illustrative benchmark produced 98.96% classification accuracy, 8.86 ms mean detection time, 1.54% mean location error, and no primary-backup coordination violation. The framework therefore provides a compact, transparent basis for later implementation in MATLAB/Simulink, PSCAD, or a relay hardware-in-the-loop platform.