Design Optimization of Savonius Wind Turbine Using CFD-Particle Swarm Optimization with Power Flow Validation Experimentally

Abstract

Air wheel generator or wind turbine generator has capabilities in generating electricity even at the lowest speed. Recently, the Savonius wind turbine with combined wind deflector and housing that applied at the top of car roof. In this paper, the unidirectional vertical axis wind turbine with two blades of Savonius type and belting system have been numerically optimizes using Computational-Particle Swarm Optimization (CFD-PSO). The 5-step optimization was done systematically to determine the related formulation based on the geometrical dimension of wind turbine blade and design constraints. The simplified model of the classical cantilever beam was used in formulating the bending stress and deflection as a constraint. Then the simulation analysis in Matlab was performed using Particle Swarm Optimization (PSO) method in minimization the mass of blades.  The current generation was compared to the initial design OPTION 1 for the blades with a mass of 1.3414 kg. From simulation analysis results, the researcher found OPTION 3 has m3 =0.8387 kg with 37.5% mass reduction of blades. The result of OPTION 3= [0.1 0.001] with belting unit were proposed to be used in prototype development that achieving the objective in minimization the mass of blades but not fulfil the minimum power and current generated caused by the existing generator type not enough rotational speed with minimum of 1400 for cut in speed for standard automotive application.