A safe-by-design approach based on safe Bayesian optimization for tuning individual pitch controllers on floating offshore wind turbines

Individual pitch control (IPC) effectively mitigates the structural loads on floating offshore wind turbines (FOWTs). However, conventional optimization methods for IPC tuning can explore unsafe parameters. To alleviate this critical safety risk, this paper proposes a safe-by-design framework based on safe Bayesian optimization (SafeBO) for automated tuning of a proportional-integral-based individual pitch controller. Unlike standard optimization methods, the framework restricts the search to regions remaining above a predefined safety threshold, guaranteeing zero safety violations across all tested scenarios and defined metrics. The final SafeBO-tuned controller demonstrates exceptional performance. Under challenging sea states, it reduces the platform pitch motion and tower fatigue loads by 20.3% and 12.8%, respectively, from those of conventionally tuned baseline controllers, without compromising the power production. These findings confirm that the SafeBO framework effectively resolves the safety–performance tradeoff, offering a robust and practical solution for safe, autonomous optimization of FOWT control systems.