Study of Multiple Half Blades Effect on the Performance of Savonius Rotor: Experimental Study and Artificial Neural Network (ANN) Model

Abstract

Objectives: To estimate and compare the performance in terms of torque and mechanical power of a new configuration of Savonius rotors with the conventional one. Methods/Statistical Analysis: New configuration comprises multiple half blades added to conventional configuration. Two different new configurations with different half blade geometries and locations were designed. The torque and mechanical power of the rotor was measured experimentally at various wind speeds and rotor positions. The tests were done 4-6 times for each measurement and the results were averaged. Moreover, the measured data were predicted using Artificial Neural Network (ANN). Findings: The location of half blades effect the performance of the rotor. Additionally, both new configurations of Savonius rotors are associated with above 45% increase in mechanical power compared to the conventional Savonius wind turbine. Based on the simulated results, it is found that the R2 value within a range of 0.902-0.99, which indicated a very good fit of the measured data with the calculated data. Application/Improvements: ANN technique can be applied as a powerful tool and effective way in predicting and assessing the performance of the rotor (torque and mechanical power).