Design Optimization of a Vertical Axis Wind Turbine Using a Genetic Algorithm and Surrogate Models

The efficiency of vertical axis wind turbines highly depends on the aerodynamic performance of wind blades. In this paper we use modern design optimization approach to improve the wind blade performance. Design parameters include the airfoil thickness and camber. The design toolbox includes a Navier-Stokes solver, a genetic algorithm, and a response surface method. The blade performance is evaluated using a Navier-Stokes solver. A sliding mesh technique is employed to handle the rotating motion of blades. The response surface (RS) methodology, kriging method, and radial basis neural network will be applied to build a surrogate model to computationally effectively explore the relationship between design variables and power performance. A real-coded genetic algorithm (GA) is carried out to obtain optimized solutions from RS function. Our preliminary results show that increasing the blade thickness can increase the blade performance.