Responses analysis and structural optimization of a semisubmersible platform based on the Taguchi experiment method.

Due to the early stage of development for floating offshore wind turbines (FOWTs), there's a scarcity of effective methods and procedures for determining the overall dimensions of floating foundation. Moreover, the comprehensive dynamic response features of various foundation configurations remain inadequately understood. To tackle these limitations, this paper presents a global design approach for semi-submersible platform used to support a large-scale wind turbine. In this paper, the distance between offset column and main column (R OI), the radius of offset column (R O), the height of heave plate (H HP), the total draft of platform (H dft), the ratio of heave plate radius to offset column radius (R HP), and the ratio of the horizontal distance from anchor to fairlead to the water depth (x M) were selected as the initial design variables of the platform. A design of experiment based on the Taguchi method was performed to construct various combinations of design variables. Then, the physical properties and hydrodynamics of the platform, and controller dynamics were determined for each combination. The correlations between the variables and key outputs were studied. Finally, the multi-objective optimization for the foundation dimensions was performed based on the Taguchi method. Compared to other experimental cases, the FOWT with optimal variables (R OI = 29 m, R HP = 2.0, R O = 7 m, H HP = 6.0 m, H dft = 21 m, and x M = 4.0) exhibits excellent comprehensive performance, particularly outperforming the OC4 model overall, with a 26% increase in platform mass (excluding ballast mass). This study offers an insight into the nature of the design variables and the response outputs, and provides a optimization framework to optimize the foundation dimensions, which can then serve as initial points for a more detailed design. • Natural frequency and output of a FOWT with various platform parameters were studied. • Taguchi method is used to perform the DOE. • The correlations between variables and key outputs were studied. • The multi-objective optimization of the platform structure was performed and validated. [ABSTRACT FROM AUTHOR]