Conditionally random perturbation-based adaptive MPC for suppressing inter-area oscillations of power systems integrated with renewable energy

This paper proposes an adaptive model predictive control (AMPC) to suppress the inter-area power oscillations for power systems whose operating points vary over wide ranges due to renewable power generation. The adaptive model predictive control adopts a novel logic in which the online identifying and controlling actions are alternatively and periodically implemented. Specifically, the random perturbation serving the online identification of system model is elaborated to satisfy certain conditions so that the strength of the perturbation can be fully exerted without affecting the stability of the post-perturbation system even when it is open-loop unstable and the control input is bounded. Moreover, a linear programming-based technique is proposed to online compute the required condition for the random perturbation. Numerical simulation and hard-ware-in-the-loop experiments validate the model tracking capability of the adaptive model predictive control and demonstrate its practical adaptive control capability in terms of real-time stabilization of slowly time-varying power systems suffering from inter-area oscillations.