Unified Smith predictor compensation and optimal damping control for time-delay power system

Abstract

Wide-area signals from wide-area damping controllers have non-negligible delays during acquisition and transmission in power systems. Such delays can undermine the damping performance of wide-area power system stabilizers. To handle this type of delay, we propose an optimal wide-area damping controller considering time delay. First, the open-loop power system is linearized, and the installation locations of the controller and feedback signal are selected from geometric measurements. Then, frequency-domain identification allows to obtain the system estimation model, and a unified Smith predictor provides compensation of time delay, thus mitigating the influence of time delay on the control system. Finally, particle swarm optimization is applied to further improve the controller. The proposed controller not only eliminates the influence of time delay but also ensures robustness by automatically performing particle swarm optimization and using collaborative search of individual local and group global information to speed up convergence. Taking the two-area four-machine and New England systems for testing, simulation results show that the proposed method enhances the ability of the closed-loop system to withstand time delay and effectively improves both the damping of low-frequency oscillations and the dynamic stability of the power systems.