Solar Radiation Forecasting in Saudi Arabia Using Machine Learning

Solar energy is a promising renewable energy source due to its availability and environmental friendliness. The capability to maximize the utilization and efficiency of solar energy remains a difficult task because of the challenges in collecting and analyzing solar radiation data. Therefore, there is a great need to forecast solar radiation to predict the output power. This paper aims to use machine learning methods to forecast solar radiation in Saudi Arabia in Riyadh city. The study compares the forecasted solar radiation using different machine learning models such as Artificial Neural Networks (ANN), Random Forest, and linear regression. The weather dataset was obtained from KACARE. The proposed models were evaluated using root mean square error and direction accuracy. The Random forest has the highest accuracy and the lowest RMSE with an accuracy of 92.8571, and RMSE of 10.3157 compared to ANN (91.3043, 18.4656), linear regression (78.5714, 30.4098).