A complete scheme for fault detection, classification and location in transmission lines using neural networks

Abstract

This work presents an artificial neural network (ANN) approach to simulate a complete scheme for distance protection of a transmission line. In order to perform this simulation, the distance protection task was subdivided into different neural network modules for fault detection, fault classification as well as fault location in different protection zones. A complete integration amongst these different modules is then essential for the correct behaviour of the proposed technique. The three-phase voltages and currents sampled at 1 kHz, in pre and post-fault conditions, were utilised as inputs for the proposed scheme. The Alternative Transients Program (ATP) software was used to generate data for a 400 kV transmission line in a faulted condition. The NeuralWorks software was used to set up the ANN topology, train it and obtain the weights as an output. The NeuralWorks software provides a flexible environment for research and the application of techniques involving ANNs. Moreover, the supervised backpropagation algorithm was utilised during the training process.