HYBRID MODELING APPROACH FOR DESIGNING VARIABLE TWIST BLADES: IMPLEMENTING CFD AND DATA DRIVEN MODELS FOR OCEAN CURRENT TURBINE TWIST OPTIMIZATION

The paper below discusses a design approach for Ocean Current Turbines (OCTs) with variable twist blades for improved energy capture and reduced loading. This paper focuses on the intersection of two key methodologies for simulating rotor behavior. The first methodology is the OpenFAST models used to optimize, and the second simulation method is CFD for detailed flow overview. The OpenFAST software developed by the National Renewable Energy Laboratory (NREL) is used in conjunction with Aerodyne to produce the turbine characteristics at various speeds; this is used for baseline results. The CFD setup uses a virtual disk model with Large Eddy Simulation (LES) to capture the detailed wake dynamics of the flow field. Simulations were employing STAR-CCM+ to capture wake evolution and flow recovery trends, validating the potential benefits of adaptive rotor strategies. This paper also covers a hybrid modeling methodology currently being tested to interface the optimization techniques and CFD models. This methodology uses a data-driven model with results from the CFD simulations. A genetic algorithm was used to optimize the TAD; it showed a reduction in fatigue loading of 23.09%. Future work will focus on further refinement of the optimization framework and its integration with CFD.