Development of a control co-design optimization framework with aeroelastic-control coupling for floating offshore wind turbines

Abstract

Sequential design optimization is frequently used to design floating offshore wind turbines. This method simplifies the design problem by implementing each of the disciplines successively and independently with the direct/dynamic coupling between them neglected. However, if the coupling between disciplines is strong, the global optimum may not be found by sequential design. Control co-design optimization is an approach that integrates all physics in a single optimization formulation to achieve the global optimum. However, it is challenging to achieve due to the highly complex system-control coupling in floating offshore wind turbines. This paper describes a control co-design optimization framework developed for floating offshore wind turbines. In particular, the blade airfoil chord and twist distributions are parameterized to change aerodynamics while the blade structural characteristics can be updated. The rotor speed and blade pitch controllers are tuned automatically based on the properties of the new plant model. For demonstration, a 10 MW reference wind turbine is used in the case study. The annual energy production is maximized, and the blade mass is minimized. The blade shape and control parameters are optimized simultaneously, subject to constraints on geometric design variables, fatigue loads, ultimate loads, and blade deflection. To solve this high-dimensional, nonlinear, and constrained problem, a genetic algorithm is used, and the Pareto front of the design space is found. The design solutions on the Pareto front showed an improvement in the blade mass and annual energy production. These design alternatives also showed reduced structural ultimate and fatigue loads, and generator torque. This revealed the potential of control co-design to reach a better performance than the sequential method.