A method of data mining for selection of site for wind turbines

The paper aims at proposing a data mining framework which will help in selection of suitable site for wind turbine׳s installation. After thorough the literature review, a list of indicators is prepared which is used for analysis of a particular site consisting of wind speed, different built-ups, and distance of existing energy installations from the prospect site, cost dependent factors, and ecological impacts. These dimensions caused increase in the attributes of dataset which are then reduced by using principal component analysis. The resulting components are regressed by using multiple regression technique. These techniques are applied on an integrated database which is prepared by storing the data of 39 sites in Pakistan. The prediction of the model developed for the wind energy site has been found to be significantly accurate when compared with expert opinion and previous studies.