Topology processing and static state estimation using artificial neural networks

Abstract

The paper presents a new approach based on artificial neural networks (ANN) for power system network topology determination and static state estimation. The state estimator model considers dynamic change in network topology and bad data processing. ANN models based on the counterpropagation network (CPN) and functional link network (FLN) have been tried out for solving topology processing and static state estimation on the IEEE 14-bus, IEEE 57-bus and a 19-bus practical Indian system. In addition the results of state estimation using the Hopfield neural network and the conventional fast decoupled state estimator (FDSE) have also been obtained and compared. It has been established that the ANN based models provide results much faster, and work well even for non-Gaussian noise and in the presence of bad data.