Indirect symmetrical PST protection based on phase angle shift and optimal radial basis function neural network

This paper proposes a new algorithm for blocking the operation of Indirect Symmetrical Phase Shift Transformer (ISPST) differential relay when subjected to different operating conditions except internal fault condition. The proposed algorithm is amalgamation of Phase angle shift (PAS) threshold and Optimal Radial Basis Function Neural Network (ORBFNN). PAS between source and load side currents of fundamental frequency is used as threshold. PAS threshold identifies whether the abnormal condition is in ISPST or out of ISPST. ORBFNN is used for the discrimination of internal fault from magnetizing inrush condition. The ORBFNN is designed by using Particle swarm optimization (PSO) technique. The performance of proposed ORBFNN algorithm is compared with more commonly reported Feed Forward Back Propagation Neural Network (FFBPNN). The simulations of different operating conditions of an ISPST are performed by using PSCAD/EMTDC software.