Potential Assessment of Solar Photovoltaic Generation using Machine Learning Algorithms for Southern Region of India

Main Article Content

P. Upendra Kumar, K. Lakshmana Rao, T. S. Kishore


Now a day, large scale grid interconnected SPV generation systems are increasing day by day, the stable operation of grid highly depends on the amount of SPV energy penetrating into the grid. This is not only essential for stable operation but also necessary for generation allocation and load scheduling. In order to achieve this, a precise method for estimating the potential is necessary. In this paper, a modest attempt has been made to estimate the potential of SPV generation for southern region of India. The methodology presented is based on an efficient machine learning algorithmbased regression methods viz. logistic, linear, support vector and random forest regression models for prediction of number of units’ generated has been presented.To evaluate the efficiency of these algorithms key performance indicators such as mean absolute error, mean square error, root mean square error and R2 score have been considered. It has been observed that random forest model performs better than all the other methods considered in this study and the same was summarized in the results.

Article Details