Dynamic Event-Based Model Predictive Load Frequency Control for Power Systems Under Cyber Attacks

Abstract

This article is concerned with the load frequency control (LFC) for the multi-area power system (MAPS) in the presence of cyber-attacks and disturbances. A resilient model predictive control (RMPC) strategy combined with the mixed $H_{2} /H_{\infty } $ performance index and dynamic event-triggered mechanism (DETM) is developed by using the cone complementarity linearization algorithm and the linear matrix inequality technique. With the application of $H_{2} /H_{\infty } $ performance index, the resilience of the controlled MAPS is achieved and the cyber-attacks launched on the MAPS can be dealt with effectively. The goal of the DETM is to enhance the communication efficiency by interrupting the communication of the MAPS, the triggering condition of which utilizes the internal dynamic variable. Moreover, rigorous derivations on the recursive feasibility of the RMPC are given. Finally, a MAPS is given to illustrate the validity of the presented control strategy.