Analysis of power generation efficiency of large Indian wind farms using computational fluid dynamics and polynomial regression

Published: 17 February 2019| Version 1 | DOI: 10.17632/y4k226y3hw.1
Nimish Mishra


The Deccan coastline in India has a great potential for wind power generation, due to balanced wind speed spread over a long period of time. Most of the large wind farms (employing large wind turbines) in India employ state-of-the art wind turbines and are meeting the energy demands put upon them. Other sites, however, still use old and inefficient turbines and are not able to exploit their full potential. India’s energy demand is set to increase in the upcoming years. It is essential, therefore, to revamp old designs. This dataset is a part of an attempt that aims to estimate the farms’ productivity by simulating the GE 1.5XLE turbines in these areas. Simulation experiments were done in ANSYS to record the simulation data. A polynomial regression model was trained on this data in order to predict the estimated power output for any value of the input wind velocity. The results were then compared with existing data about the actual power output of the wind farms. Results were documented. References for existing data: Windpower (Wind energy market intelligence). Lalpur wind farm report. [Last checked February 16, 2019] Windpower (Wind energy market intelligence). Indian wind farms data. [Last checked February 16, 2019] References for wind velocity data: a. Sandeep Chinta, Arun Agarwal, C. Raghavendra Rao. Wind Speed Model for Anantapur District, Western Andhra Pradesh, India. International Journal of Innovative Research in Advanced Engineering (IJIRAE) [2014] b. World Weather Online, Monthly Climate Averages [Last checked February 16, 2019] c. Indiastat; Indiastat Month Wise Mean Wind Speed [Last checked February 16, 2019] d. Meteoblue Weather history: [Last checked February 16, 2019]



Computational Fluid Dynamics, Linear Regression, Computational Methods in Fluid Dynamics