Discrimination Between Inrush and Internal Fault Currents in Protection Based Power Transformer using DWT

Inrush currents in electric power transformers are nonsinusoidal, high currents in magnitude that are generated due to saturation of flux in the transformer core during the energization.This paper proposes a decision technique for discrimination between inrush and internal faults currents in power transformers.The technique based on wave shapes recognition by criterion extracted of the instantaneous differential currents using the discrete wavelet transform based feature extraction technique.Adaptive Neuro-Fuzzy Inference System is used as a machine learning technique to classify inrush current and internal fault current of a power transformer.The criteria based on the calculation of energy contents in extracting coefficients D4 and D5 fed to ANFIS controller to give the decision.Simulation and lab tests have been implemented.The experimental results show the discrimination can be accurately made by quantifying the extracted features, also shows the technique is reliable, quite fast, and appropriate for different types of the power transformer.