Passive flow control in lift-type vertical-axis wind turbines using optimized flaps

This study demonstrates the passive control of airflow through lift-type vertical-axis wind turbines (VAWTs) by incorporating flaps on the rear of the blades. The objective is to enhance aerodynamic performance, particularly the power coefficient, by optimizing the flap configuration to its fullest potential. We employ high-fidelity two-dimensional unsteady Reynolds-averaged Navier–Stokes simulations with the transition shear stress transport turbulence model to examine the impact of flap length, angle, and chordwise location on turbine performance during dynamic stall. The simulation strategy is founded on a Taguchi experimental design. A machine learning (ML) model utilizing an artificial neural network is developed on computational fluid dynamics (CFD) results to comprehend the nonlinear correlation between flap parameters and power output. This reduces the expense of performing numerous simulations. This surrogate is integrated with a covariance matrix adaptation evolution strategy to efficiently identify optimal flap configurations. The integrated CFD-ML–optimization framework demonstrates that optimally designed flaps can significantly enhance the turbine's power coefficient by altering the patterns of local flow separation and vorticity. The proposed method enhances the aerodynamics of VAWTs and is both computationally efficient and conceptually clear in its physical principles. This exemplifies a surrogate-assisted optimization study centered on passive flaps in VAWTs.