On-line dynamic security assessment using a real-time power system simulator with neural network contingency screens

On-line dynamic security assessment aims to provide power system operators with real-time information on the stability of a power system subjected to a set of probable contingencies. Although this task is several orders of magnitude more complex than static security analysis, recent advancements In computer technology now make such analysis possible within the time constraints required by an energy management system. This paper describes a dynamic security assessment system that has recently been developed and provides details of the artificial neural network based stability screens that are used to improve the performance of the system. Simulation results are presented for various snapshots of the UK national grid system which show that full on-line dynamic security assessment is now practical.