Session: 06-11-02 Ocean Engineering Technology - II

Submission Number: 180458

Data Driven Individual Pitch Control for FOWTS: A Gaussian Process and Cross-Entropy MPC Approach 

Floating offshore wind turbines (FOWTs) have emerged as a promising solution for deep-water wind energy harvesting. However, they suffer from significant structural fatigue loads caused by the coupled aero-hydro-servo-elastic dynamics. These loads pose a major challenge to the reliability and cost-effectiveness of FOWTs. To address this issue, this paper proposes a data-driven nonlinear model predictive control (MPC) framework for individual pitch control (IPC) to mitigate asymmetric fatigue loads. A Gaussian process (GP) regression model is first trained on high-fidelity simulation data to capture the complex nonlinear dynamics of the FOWT. The probabilistic nature of the GP provides accurate predictions while quantifying model uncertainty, which supports more robust control design. Based on this model, a nonlinear MPC problem is formulated and solved online using the cross-entropy method (CEM), a derivative-free stochastic optimization algorithm. Compared with conventional gradient-based solvers, CEM efficiently handles the nonconvex and nondifferentiable nature of the control problem by iteratively updating elite samples toward the global optimum. The proposed GP-CEM-MPC controller is validated in the OpenFAST simulation environment and benchmarked against a conventional baseline controller. Simulation results under turbulent wind and irregular wave conditions show that the proposed approach significantly reduces fatigue loads, measured by damage equivalent loads (DELs), on critical components such as the blade roots and tower base. Meanwhile, it maintains rated power output and satisfies physical constraints. The results confirm that this control strategy is a practical and effective approach to enhance the structural longevity and performance of FOWTs.

Presenting Author: Yanhua Liu Shandong University

Presenting Author Biography: Yanhua Liu received the B.Eng., M.Eng. degrees in Automation department from the North China Electric Power
University, China, in 2012 and 2015, respectively. She achieved her Ph.D. degree in Control & Intelligent Systems
Engineering Research Group, University of Hull, in 2019. From 2019 to 2021, She worked as a postdoc in the project ‘‘A New Partnership in Offshore Wind’’, Department of Engineering, University of Hull. She is currently a Lecturer with the School of Electrical Engineering, Shandong University, Jinan, China. Her research interests lie in the power electronics, offshore wind turbine, wave energy and model predictive control.

Authors:

Aolong Fu Shandong University
Xiuzhong Gong Huadian Electric Power Research Institute Co.,LTD
Yanhua Liu Shandong University
Ziyang Han Shandong University
Shuo Shi State Grid Shandong Electric Power Research Institute