An Adaptive Distance Relaying Scheme Using Radial Basis Function Neural Network

Abstract

The distance calculation performed by a distance relay is incorrect due to ground fault resistance, prefault system conditions, mutual coupling effect and shunt capacitance influences. The problem is more serious keeping in view that all relay settings are made based on some compromise. The work presented in this article addresses the problems encountered by conventional pilot independent distance relay when protecting two terminal transmission lines. The key feature is that it presents the detailed analysis of the apparent impedance as seen from the relaying point taking into account the effects of transmission line parameter uncertainties, shunt capacitance influences and variations in the system external to the protected line. These results can be used for estimation of correct impedance of transmission line, hence taking relaying decisions. Based on the results of extensive computer simulation of the infeed/outfeed, fault resistance and shunt capacitance effects on the relay characteristics, an adaptive distance relaying scheme is proposed for such lines using a radial basis function neural network (RBFNN), which provides a more efficient approach for training, computation, adaptation, and tripping than the conventional feed forward network using back propagation algorithm (BPNN). The proposed adaptive protection scheme is tested for a single line to ground fault, but for varying fault locations, fault resistances, fault inception angles, and different source impedance ratio too. At the end, comparison of conventional feed forward network using back propagation algorithm with radial basis function classifier is also given, which clearly indicates improvement. Hence, the selectivity of the protection system is increased, as is the power system reliability.