STEADY-STATE SECURITY EVALUATION OF ELECTRICAL-POWER SYSTEMS BY MEANS OF ARTIFICIAL NEURAL-NETWORK

An artificial-neural-network (ANN) application for steady-state security evaluation of electrical power systems is presented. Such application is based upon a combined use of a multilayer back-propagation neural network and of an opportune electrical system model, previously proposed by some of the authors. In the paper; the definition of steady-state security and the problems arising from its on-line evaluation are briefly recalled Then, the ANN utilized is described with particular reference to the solutions adopted by the authors. Afterwards, some remarks on the system model utilized are reported. Finally, numerical applications on the AEP-14 bus system are presented and the results obtained with different ANN configurations are analyzed.