Identification of type of internal fault in indirect symmetrical phase shift transformer based on PRN

Abstract

This paper describes a technique for the detection of type of internal fault in an indirect symmetrical phase shift transformer (ISPST). An application of Pattern Recognition Network (PRN) is proposed as a core classifier to identify the type of internal fault. Four type of internal faults (turn-to-turn (TT), line-to-ground (LG), two line-to-ground (LLG), and three line-to-ground (LLLG)) have been classified. Numerous test cases of internal fault in an ISPST have been using PSCAD/EMTDC software. These cases are formed on the basic variation of different parameters of ISPST like fault inception angle, fault resistance loading condition and percentage of winding. The accuracy of the proposed technique is evaluated over a large number of cases and it is observed that the technique gives the results with high accuracy even in presence of noise in the signal.