IMPROVING CONTROL PERFORMANCES OF AN LQR-BASED CONTROL STRATEGY: A BIOINSPIRED OPTIMIZATION APPROACH

Linear quadratic regulation (LQR) has been treated as one of the most popular control strategies of a control system. Technically, this control methodology is based upon solving a special equation named Riccati algebraic equation to determine a decision vector, then to design the full state feedback control law. In such a control strategy, it is indispensable to decide two weighting matrices Q and R which can strongly affect the control performances of a system under consideration. This paper proposes a novel method to optimally determine the values for all elements of the matrices Q and R. This is executed by means of an efficient bio-inspired optimization mechanism. Whenever the two matrices Q and R are updated following the optimization technique, the state vector will be improved and a fitness or cost function is always evaluated to optimize the control performance of the system. The proposed control strategy will also be demonstrated through a typical case study of load-frequency control problem of a thermal power plant. A significant number of numerical simulation results will be provided to testify the feasibility and superiority of the power system applying the LQR scheme proposed in this study over the conventional PID regulators.