Potential Classification Prediction of Solar and Wind Energy in Indonesia using Machine Learning with Random Forest Algorithm

The realization of the mass use of renewable energy needs to be supported by various parties, including Indonesia, which is a country that has great potential in the use of renewable energy. The community, government, and private parties who become developers in this regard need to realize the main potential of regions in Indonesia. This study aims to determine the prediction results of renewable energy in Indonesia, especially the potential for solar and wind energy. The method used in this research is machine learning with a random forest algorithm. Random forest classification prediction modeling was carried out using R software. The data samples used were data on maximum temperature, minimum temperature, average temperature, average humidity, duration of sunshine, and average wind speed from 34 provinces in Indonesia. Based on the prediction results with random forest, the average accuracy of the random forest classification model ranges from 81-92%. So, it can be said that the model can provide reasonable predictions. Based on random forest predictions, the classification of the largest level of solar energy potential is in 3 main areas, namely the provinces of East N usa Tenggara, Bali, and West Nusa Tenggara. While the areas with the greatest wind energy potential are in 3 provinces, namely Kep. Bangka Belitung, East Nusa Tenggara, and East Java. The province of East Nusa Tenggara has a high potential for solar and wind energy. The potential duration of solar radiation reaches 10.6 hours per day, while the average wind speed reaches 4 m/s per day. The results of this study are expected to be input for parties who play a role in developing renewable energy in Indonesia to support the renewable energy mix target of 23 % by 2025.