ANN application techniques for power system stability estimation

The implementation of artificial neural networks (ANN) as a power system st ability monitoring tool is a viable option, introducing dynamic and intelli gent solution to utility operators. This paper examines the performance of two nonlinear multilayer ANN models which are similar in structural topolog y and training emphasis but different by way of the utilization of their ne t or basis function. The performance of both models were compared for the e stimation of stability index to gauge the stability of a power system netwo rk. Although tests were conducted in a simulated environment, loading patte rns analyzed in this case study were realistically generated, and hence tes t results realistically accentuates the potential of ANN for practical on-l ine dynamic system implementation.