Gradient-based model predictive control for energy capture optimization of oscillating buoy wave energy converters

The appropriate application of model predictive control (MPC) for oscillating buoy wave energy converters (OBWEC) can significantly improve energy capture efficiency while ensuring safe operation. In this paper, a gradient-based MPC, demonstrated to have high computational efficiency, is proposed for OBWEC. The state space model of buoy is established based on the Cummins theory, and wave prediction is achieved through deterministic sea wave prediction method. The optimal control problem of MPC is formulated as a two-point boundary value problem and solved using the gradient projection algorithm with adaptive line search method. To handle the constraints associated with the physical limitations of the power-take-off system, the augmented Lagrangian approach is implemented. For benchmarking purposes, passive control is utilized for algorithm parameter tuning under regular wave conditions. The proposed controller performance under irregular wave conditions with prediction errors is subsequently evaluated against an MPC approach formulated as a nonlinear programming problem and solved via sequential quadratic programming. Comparative results demonstrate that the proposed gradient-based MPC method achieves a notable improvement in energy capture efficiency relative to these benchmark strategies.