A novel adaptive model predictive frequency control using unscented Kalman filter

This paper proposes a novel adaptive load frequency control (LFC) method based on model predictive control (MPC) using an unscented Kalman filter (UKF). The objective is to develop a robust LFC controller against the change in frequency characteristics of a power system. An adaptive MPC (AMPC) control is realized by updating the prediction model on-line to capture the frequency characteristics of a target system. A simplified prediction model is identified in the preliminary examination of the closed-loop performance of the target system, and then the system parameters are estimated in real-time circumstances. The estimated parameters include equivalent inertia of the target system. An interconnected three-area test system is used in the simulations to confirm the effectiveness of the proposed approach. The simulation results verify that the proposed MPC method can work successfully and efficiently handle the disturbances of the power system. The successful results also confirm that the proposed control scheme can be applied to a microgrid (MG) project, which is being promoted by the authors.