Prediction of daily global solar radiation and air temperature using six machine learning algorithms; a case of 27 European countries

The prediction of global solar radiation in a region is great importance as it provides investors and politicians with more detailed knowledge about the resource that region, which can be very beneficial for large-scale energy development. In this sense, main objective study to predict daily data 27 cities (Brussels, Paris, Lisbon, Madrid…), located countries, have mostly different distributions Europe. research, six machine-learning algorithms (Linear model (LM), Decision Tree (DT), Support Vector Machine (SVM), Deep Learning (DL), Random Forest (RF) Gradient Boosted Trees (GBT)) are used. training these algorithms, air temperature(Ta), wind speed(Va), relative humidity(RH) supplied from Meteonorm tool cover last years grouped two periods (1960–1990; 2000–2019). To decide on success four statistical metrics (Average Relative Error (ARE), Average absolute (AAE), Root Mean Squared (RMSE), R 2 (R-Squared)) discussed study. addition, forecasting temperature 2050 2100 were made using three most recent Intergovernmental Panel Climate Change (IPCC) scenarios (RCP2.6; RCP 4.5, 8.5). results show ARE, R, RMSE values all ranging 0.114 6.321, 0.382 0.985, 0.145 2.126 MJ/m , respectively. By analysing noticed tree exhibited worst result terms metrics. Among DL was recognized only algorithm exceeded t-critical value (The cutoff between retaining or rejecting null hypothesis). Globally, machine learning used research applied good accuracy. Despite this, SVM best among models It followed by DL, LM, GB, RF DT, • R2, metrics, t-critic value, Algorithms shown successful