Prediction Of Solar Power Generation Based On Machine Learning Algorithm

Abstract

Energy demand is growing and by 2050 solar energy will account for 11% total electricity production. It has emerged as one of the most potential sources of alternative energy Even though the usage of solar energy in residential places has increased, yet they are regarded as unpredictable and irregular power sources because the generated power output depends on the geographical region, atmospheric conditions, which can vary drastically. Depending upon the weather conditions solar panels will work differently. Since the power generation mostly depends on weather conditions it is necessary to consider weather conditions. Because of the unpredictability in photovoltaic generations, it is crucial to examine the effects of environmental circumstances on solar power system using machine learning based approach. The machine learning algorithm shows great results in anticipating the power with weather conditions as input models. The approach uses different databases, input, and mathematical relationships to predict the solar power generated. Various machine learning algorithm would be applied to get the patterns and to obtain the results with maximum accuracy and efficiency. This study demonstrates how a variety of machine learning techniques may be used to predict the amount of energy a solar panel provides. Various models were applied to the database and the most appropriate machine learning predictive model was identified through coefficient of determination analysis. The results obtained after comparing the data for different years are furnished. Temperature, relative humidity, pressure, and wind speed are the independent factors, with power generated as the dependent variable. The proposed model has provided prediction results with good accuracy.