Efficiency improvement of a vertical-axis wind turbine using a deflector optimized by Taguchi approach with modified additive method

[Display omitted] •A deflector optimal design via a modified Taguchi approach is developed.•The influential extents of design factors are analyzed by analysis of variance.•The VAWT with the optimized upper deflector can improve performance by 20%.•The VAWT with the optimized lower deflector can improve performance by 17%.•The optimal combinations of factors are accurately predicted by the modified model. This study designs a deflector to enhance the performance of a three-bladed vertical axis wind turbine (VAWT) through the Taguchi method in association with a modified additive model (MAM). Two types of the deflector, namely, an upper deflector and a lower deflector, are individually considered. Analysis of variance (ANOVA) is applied to evaluate the influences of design factors on the performance, and the aerodynamic characteristics around the turbine blades are explored. The prediction suggests that the average power coefficients (Cp-) of the VAWT without deflector is 0.46. For the upper deflector design, the Cp- value using MAM can be raised to the highest value of 0.55, while the Cp- value for the lower deflector design is lifted to 0.54, accounting for 20% and 17% improvements compared with the VAWT without a deflector. The application of MAM shows that the distances from the center of the turbine to the bottom of the upper deflector and the top of the lower deflector are the most important factors to influence Cp-, and the interaction between the factors should be considered to provide accurate optimization results. The results of the ANOVA analysis are coincident with those of the Taguchi approach and confirm that a proper installation of a deflector can efficiently intensify the VAWT’s performance.