Optimal design of Savonius wind turbine blade based on support vector regression surrogate model and modified flower pollination algorithm. (15th October 2022)

• Record Type: Journal Article
• Authors: Jia, Rongyuan; Xia, Huaijie; Zhang, Song; Su, Weiguang; Xu, Shuhui
• Is Part Of: Energy conversion and management. Volume 270(2022)
• Publication Date: 2022-10-15
• Subjects: Savonius wind turbine -- Bezier curves -- Latin hypercube sampling -- CFD simulation -- SVR model -- Modified flower pollination algorithm; Direct energy conversion -- Periodicals; Energy storage -- Periodicals; Energy transfer -- Periodicals; Énergie -- Conversion directe -- Périodiques; Direct energy conversion; Periodicals; 621.3105
• DOI: 10.1016/j.enconman.2022.116247
• Languages: English
• ISSNs: 0196-8904

Abstract

Highlights: The cubic Bezier curve is used to describe the shape of the blade. SVR model is used to catch the relationship of the shape and performance. The Modified flower pollination algorithm is used to get the final design scheme. The performance of the optimized blade is notably better than the classical blade. Abstract: Blade shape has a significant effect on the wind-capturing ability of the Savonius wind turbine. This paper proposes an intelligent optimization method for optimizing the shape of the blade of the Savonius wind turbine. This method firstly uses the cubic Bezier curve with four design parameters ( x 1, y 1, x 2, y 2 ) to characterize the complex blade shape, and then uses the Latin hypercube sampling method to sample some design schemes throughout the design space, and uses computational fluid dynamics simulation to evaluate the response value, i.e. the moment coefficient, of each scheme, and then uses the support vector regression surrogate model to describe the relationship between the design parameters and their response values, finally uses modified flower pollination algorithm to solve the surrogate model to obtain the optimal blade shape. After comparing and analyzing the optimized blade and the classical semicircular blade, it is found that compared with the classical semicircular blade, the optimized blade has a better wind-capturing ability. When the wind speed is 7 m/s and the tip speed ratio is 1, its average power coefficient Cp is significantly increased from 0.260027 to 0.277902 (about 6.87% higher). In addition, the aerodynamic performance of the optimized blade is also better than the classical semicircular blade at other tip speed ratios ( TSRs = 0.6–1.2). It is shown that the wind turbine with the optimized blade has great potential in a practical application environment.