The usage of 10-fold cross-validation and grid search to enhance ML methods performance in solar farm power generation prediction
Seyed Matin Malakouti, Mohammad Bagher Menhaj, Amir Abolfazl Suratgar
Abstract
It is essential to have accurate projections of the quantity of solar energy that will be generated in the future to improve the competitiveness of solar power plants in the energy market and reduce the dependence of both the economy and society on fossil fuels. This can be accomplished by having a better understanding of the amount of solar energy that will be generated in the future. We used databases containing information about California that span 2019 through 2021. These years encompass the state's forecast. These data were used in the analysis. The 10-fold cross-validation and Grid search has been used to enhance the performance of decision tree, light gradient boosting machine, and an extra tree in Solar Farm Power Generation Prediction.