A sensorless model predictive control framework for wave energy converters utilizing extended Kalman filter-based excitation force estimation

Wave energy converters (WECs) face significant challenges in control performance due to system nonlinearity, dynamic parameter coupling, and reliance on costly physical sensors. To address these issues, this paper proposes a sensorless control strategy that combines an Extended Kalman Filter (EKF)-based excitation force estimation with Model Predictive Control (MPC). The EKF estimates excitation forces in real time using measurable buoy motion states—namely, heave velocity and displacement—eliminating the need for direct wave force measurements. The estimated forces are then integrated into an MPC framework to address parameter coupling among hydrodynamics, power take-off (PTO) damping, and generator load. Simulation results demonstrate that the proposed method achieves comparable control accuracy to conventional MPC strategies based on known excitation inputs, while significantly reducing system complexity and sensor costs.