Skip to Main Content
This study presents a new support vector machine (SVM)-based fault zone identification scheme for busbar which correctly identifies faults occurring inside and outside the protection zone of busbar. The proposed scheme utilises one cycle post-fault current signals of all the lines as an input to SVM. In order to achieve the most optimised classifier, Gaussian radial basis function has been used for training of SVM. Feasibility of the proposed scheme has been tested by modelling an existing 400 kV Indian busbar system in PSCAD/EMTDC software package. More than 28 800 fault cases with varying fault resistances, fault inception angles, fault locations, types of faults and source impedances have been generated and used for validation of the proposed scheme. The proposed scheme effectively discriminates between in-zone and out-of-zone faults with very high fault classification accuracy for different fault and system conditions. Moreover, the proposed scheme remains stable during an early and severe current transformer (CT) saturation condition giving an accuracy of 99 for all the fault cases.
Date of Publication: October 2011