Use of artificial neural networks in a dispatcher training simulator for power system dynamic security assessment

Preliminary results on using artificial neural networks (ANNs) as a dispatcher's aid for power system dynamic security assessment are presented. Dynamic security assessment is explained, and a brief introduction to ANNs is given. The ANN-based dispatcher training simulator is then described, and results for a 39-bus test system are given. It is found that the ANN is able to properly estimate the postfault state and accurately predict the security status of the power system. In general, the ANN results are in close agreement with those obtained by simulation.<
>