Identification of Geographical Locations to Operate Savonius Wind Turbine Rotor for Meeting a Desired Performance

In this paper, feasible geographical locations in India have been identified to meet a desired performance criterion from a Savonius wind turbine rotor involving semicircular blades. The identification is based upon the average wind speed prevailing at the relevant location. For a given turbine geometry, in order to simultaneously satisfy the required power and torque characteristics over a particular range of tip speed ratio, an inverse problem is solved with the aid of golden section search method (GSSM)-based optimization algorithm to predict the required local wind speed. For this, the minimization of the sum of least square errors between the target power and torque coefficients is done with respect to some initially-guessed power and torque values. Thereafter, based on the estimated wind speed, the reconstructed power and torque characteristic curves are validated with the experimental wind tunnel data. The necessary blockage corrections have been considered during the inverse analysis for which pertinent correlations reported in the available literature are used. The variations of the estimated parameter and the pertinent objective function are studied at different iterations of the GSSM. The effect of the initial guess on the estimated value of wind velocity is also reported and it is found that a unique solution occurs for a particular set of power and torque characteristics. The present work avoids the conventional hit and trial method based nonlinear analysis along with repetitive field tests which are otherwise needed to simultaneously generate a given power and torque performance from the Savonius wind turbine. The proposed inverse method thus can be extremely useful to determine the feasible Indian geographical locations directly from any required torque and power data.