Application of Support Vector Machine for digital protection of power transformer

Abstract

This paper presents an application of Support Vector Machine (SVM) for digital protection of power transformer which effectively discriminates internal faults with non-internal faults. Internal faults consist of phase to ground, phase to phase and phase to phase to ground faults, whereas non-internal faults include different types of magnetising inrush, external fault & normal condition. The existing 315 MVA power transformer of Gujarat Transmission Corporation Limited, Gujarat, India has been modelled using PSCAD/EMTDC software package. SVM classifier has been used to discriminate the internal faults of the power transformer with simulated data set of more than 4600 operating states in MATLAB. Using optimum kernel function of SVM, the overall fault classification accuracy has been compared with another digital protection technique using Probabilistic Neural Network (PNN). The result shows that SVM outperforms the PNN for discrimination of internal fault of power transformer.