Enhancing FOWT performance through GA-based control parameter optimisation: A trade-off between power and fatigue

Floating offshore wind turbines (FOWTs) are crucial for the clean energy transition. To ensure cost-effective and competitive deployment, it is essential to maximise power generation while extending operational lifetime. Environmental forces such as wind, waves, and currents induce structural fatigue, reducing lifespan. Optimising control strategies is vital, as they influence turbine dynamics, mitigate environmental loads, and enhance performance. Genetic Algorithms (GAs), inspired by biological evolution, are effective tools for optimising these strategies due to their robustness in handling complex, non-linear systems, improving both performance and durability. This study presents a methodology for optimising the Reference Open-Source Controller (ROSCO) parameters for a coupled FOWT model using GAs. The primary objective is to reduce structural fatigue without compromising power output. The approach follows a bottom-up strategy—starting with a limited set of tuning parameters and load cases, then progressively increasing complexity—to develop a comprehensive and generalisable optimisation framework. The methodology includes simulation time reduction techniques to ensure computational feasibility. Results: Show that the optimised controller achieves up to a 10.04 % reduction in tower base bending moment fatigue loads while maintaining power output within 7.40 % of the baseline. The analysis also highlights the trade-offs between control parameters and performance metrics, offering insights into their relative influence. This work contributes a flexible, scalable optimisation framework applicable to various FOWT designs and sites, with potential to reduce operational costs and extend turbine lifespan in real-world offshore wind technologies.