Discrimination of Internal Faults and Other Transients in an Interconnected System With Power Transformers and Phase Angle Regulators

This study solves the problem of accurate detection internal faults and classification transients in a 5-bus interconnected system for Phase Angle Regulators (PAR) Power Transformers. The analysis prevents mal-operation differential relays case other than which include magnetizing inrush, sympathetic external with CT saturation, capacitor switching, non-linear load ferroresonance. A gradient boosting classifier (GBC) is used to distinguish from transient disturbances based on 1.5 cycles 3-phase currents registered by change detector. After an fault, GBCs are locate faulty unit (Power Transformer, PAR series, or exciting unit) identify type fault. In disturbance detected, another GBC classifies them into six disturbances. Five most relevant frequency time domain features obtained using Information Gain train test classifiers. proposed algorithm distinguishes balanced accuracy 99.95%. transformer located 99.5% different identified 99.3%. Moreover, reliability scheme verified rating connection transformers involved, noise levels signals. These classifiers can work together conventional relay offer supervisory control over its operation. PSCAD/EMTDC software simulation develop two three-winding models creating including inter-turn inter-winding faults.