Fast Detection of Current Transformer Saturation Using Stacked Denoising Autoencoders

Malfunctions in relay protection devices are predominantly caused by current transformer (CT) saturation which produces distortion measurements and disturbances power system protection. The development of deep learning is on the rise recently because its robustness. This study presents a CT detection where secondary becomes distorted. proposed scheme offers wide range consists moving-window technique stacked denoising autoencoders. Moreover, Bayesian optimization was used to minimize difficulty determining neural network structure for approach. performance algorithm evaluated a-g faults 154 kV 345 overhead transmission line South Korea. waveform variation has been generated PSCAD different scenarios that heavily influence saturation. comparative analysis with other methods demonstrated superiority DNN method. With detect saturation, it significantly yielded high accuracy precision were approximately 99.71% 99.32%, respectively.