A Particle Swarm Optimization Method for Power System Dynamic Security Control

This paper proposes an automatic learning framework for the dynamic security control of a power system. The proposed method employs a radial basis function neural network (RBFNN), which serves to assess the dynamic security status of the power system and to estimate the effect of a corrective control action applied in the event of a disturbance. Particle swarm optimization is applied to find the optimal control action, where the objective function to be optimized is provided by the RBFNN. The method is applied on a realistic model of the Hellenic Power System and on the IEEE 50-generator test system, and its added value is shown by comparing results with the ones obtained from the application of other machine learning methods.