<![CDATA[Renewable energy plays a crucial role in reducing greenhouse gas (GHG) emissions. Excessive use of fossil fuels, such as coal, can produce GHG emissions that trigger extreme weather and global warming. Therefore, efforts to increase renewable energy utilization are necessary, in line with the Government Work Plan (RKP) target, which targets renewable energy contributions to reach 23% by 2025. This study aims to predict the potential for solar renewable energy in an area based on radiation, temperature, and rainfall variables. The method used is a supervised learning-based Random Forest. Weather data was obtained through the Open Meteo API, then processed by assigning weights to variables to produce output labels, which were then used in the classification process and model performance evaluation. The results showed that the Random Forest model produced an accuracy of 99.82%, with predictions of low/no potential energy being completely correct, medium energy potential experiencing only one error, and high energy potential also experiencing only one error. Based on these findings, the Random Forest method has proven effective in predicting solar power potential with high accuracy and is able to identify variables with the highest to lowest levels of importance.]]>
Copyrights © 2025
