Correlation Coefficients Based Fault Classification Technique for TCSC Lines

This paper presents a new fault classification technique for Thyristor-Controlled Series-Compensated (TCSC) transmission lines using features derived from correlation coefficients obtained from three phase-currents. The proposed features have been given to Random Forrest (RF) and Naïve Bayes (NB) classifiers to classify shunt faults. The proposed scheme has been evaluated in a 300 km long, 400 kV TCSC line modelled in PSCAD/EMTDC. Fault cases under diversify system conditions and fault parameters have been generated in TCSC lines. 3,840 fault cases have been used to train the classifiers. For the proposed feature, the performance of RF classifier has been found to be superior in comparison to NB classifier with fault classification accuracy of 99.65% for 14,400 fault test cases.