Applications of Hopfield neural networks to distribution feeder reconfiguration

Abstract

Distribution feeder reconfiguration is an optimization problem for loss minimization, and, in this paper, the authors investigate the use of a Hopfield neural network for distribution feeder reconfiguration. A network model is developed and presented, and then the method applied to a distribution system used by Wagner et al. (1991) consisting of three feeders, thirteen normally closed sectionalizing switches, three normally open tie switches and thirteen load points. Simulation results using this distribution system modelled as a neural network are presented.<>