Optimization of the savonius wind turbine using a genetic algorithm

This Thesis presents a methodology for an automated optimization of the rotor of a Savonius vertical axis wind turbine. This optimization was performed using an automated process integrated in a multidisciplinary design optimization software. In it, a genetic algorithm was in charge of the optimization of the selected variables. In this case the variables were the rotor profile shape, diameter and tip speed ratio of the wind turbine. This rotor's variations were evaluated by calculating its power coefficient (CP) using computational fluid dynamics (CFD). There were performed three optimizations. The first was single objective in order to maximize the CP, this was accomplished by performing modifications to the shape of the rotor profile. The second was multi objective in order to maximize the CP and minimize the difference of it in the unsteady CFD analysis (CPdif). The previous was achieved by making variations to the rotor's shape profile, size and TSR. The third optimization was single-objective (maximizing the CP) and it involved performing the same variations as the second optimization.