Application of ANN based pattern recognition technique for the protection of 3-phase power transformer

Abstract

This paper presents an application of ANN based Pattern Recognition Technique for the differential protection of a two winding three-phase power transformer. It proposes a variation in feed forward back propagation neural network (FFBPNN) model, which makes the discrimination among normal, magnetizing inrush, over-excitation and internal fault currents efficiently. Fault conditions of the transformer are simulated using MATLAB/SIMULINK in order to obtain current signals. The training process for the neural network and fault diagnosis decision are implemented using toolboxes on MATLAB. © 2011 Springer-Verlag.