A Comparitive Study on ANN Based Fault Location and Classification Technique for Double Circuit Transmission Line

Abstract

The proposed work deals with Fault classification & fault location techniques for parallel overhead transmission lines. Fault location is carried out by measuring the distributed line model of faulted line parameters. Different system faults such as LG, LLG and LLLG on a protected transmission line should be detected, classified & located rapidly in order to bring the system to the normal state. A novel application of neural network approach with three variance of feed forward neural networks such as the one with Back propagation algorithm (BPN), Radial basis function (RBF) network and Cascaded correlation feed forward network (CFBPN) is proposed for the protection of double circuit transmission line has been demonstrated in this work. The proposed method uses line current values to learn the hidden relationship in the input patterns. Using the proposed approach, fault detection, classification, location and faulted phase selection could be achieved. An improved performance is experienced once the neural network is trained sufficiently and suitably, thus performing correctly when faced with varied system parameters and conditions. Results of performance studies show that the proposed neural network-based modules outperform the performance of conventional fault selection algorithms. Among the ANN modules, result of RBF network is found to be better than the other two networks in terms of accuracy.