Optimization Of Savonius Wind Turbine Blade Based On Modified Flower Pollination Algorithm And Support Vector Regression Surrogate Model

In recent years,with the rapid reduction of non-renewable energy sources such as coal,crude oil,and gas,clean,non-polluting renewable energy sources such as wind,nuclear and tidal energy have slowly started to come into the limelight.The Savonius-type vertical axis wind turbine is a device that converts wind energy into electrical energy.This system is widely used because of its simple installation and low cost,but the low efficiency of wind capture has seriously limited its development and promotion.Therefore,to address the problem of low wind capture efficiency of this type of turbine,this thesis will conduct an in-depth study of this type of turbine by changing the shape of the turbine blades and adding auxiliary structures.The blade shape will have an important influence on the wind-catching ability of the wind turbine.Based on the summary of the literature,this thesis proposes an intelligent optimization method for the blade shape of the wind turbine.Firstly,using the blade shape characterization method proposed by our research group,i.e cubic Bessel curve,this method only needs four design parameters（x₁,y₁,x₂,y₂）to describe the complex shape of the turbine blade,then a certain number of design solutions are randomly selected in the whole design space by Latin hypercube sampling,and then the response values of moment coefficients C_m for each set of solutions are obtained by computational fluid dynamics simulation,and then support vector regression surrogate model is used to describe the relationship between design parameters and response values,Finally,the support vector regression surrogate model is solved by modified flower pollination algorithm to predict the optimal design solution for the wind turbine blade shape.Finally,a comprehensive comparative analysis of the moment,power,and flow structures of the novel blade wind turbine and the classical semi-circular blade wind turbine reveals that the novel blade has a stronger blade tip vortex and recovery flow,which helps to improve the performance of the rotor,the turbine moment coefficients improved significantly from 0.260027 to 0.277902（about 6.87%higher）at a wind speed of 7m/s and a blade tip speed ratio（TSR）of 1.The novel turbine also exhibits good wind-catching performance at other tip speed ratios（TSR=0.6～1.3）range.As this type of turbine is a resistance-based turbine,the turbine blades will be subject to wind resistance from the upstream incoming flow when returning,therefore,based on the novel blade turbine,a windshield is introduced upstream of the returning blades to reduce its negative moment.At the same time,in order to enhance the positive moment of the advancing blade,the deflector is placed upstream of the advancing blade to achieve the effect of deflecting the flow.The wind model for performance prediction was finalized after a series of validations,and the optimal geometry and placement of the windshield and deflector（turbine stator）were optimized through simulation.CFD simulation results show that using this method can improve the moment coefficients of the turbine to 0.315248,which is about 13.44%higher than the novel turbine and about21.23%higher than the classical turbine.