Application of ANN and Genetic Algorithm for Evaluation the Optimum Location of Arresters on Power Networks due to the Switching Overvoltages

Abstract

Switching surges are of primary importance in insulation co-ordination of EHV lines, as well as in designing insulation of apparatuses. The magnitude and shape of the switching overvoltages vary with the system parameters, network configuration and the point-on-wave where the switching operation takes place. This paper presents an artificial neural network (ANN) based approach to estimate the peak value of overvoltages and the global risk of failure generated by switching transients during line energizing or re-energizing in different nodes of a power network. Then a genetic algorithm (GA) based method is developed to find the best position of surge arresters on power networks so as to minimize the global risk of the network.