Prediction of Total Renewable Energy Consumption using Artificial Intelligence Approach

The increasing global energy demand and the necessity to combat climate change have underscored the importance of renewable energy sources in sustainable development. This study investigates the application of advanced Artificial Intelligence (AI) approach—Long Short-Term Memory (LSTM), Seasonal Autoregressive Integrated Moving Average (SARIMA), Random Forest, and Support Vector Machine (SVM)—to forecast renewable energy consumption. Using a monthly Total Renewable Energy Consumption dataset from January 1973 to February 2024, sourced from Our World in Data. The research aims to enhance the accuracy of consumption forecasts to aid stakeholders in optimizing resource allocation, infrastructure development, and strategic decision-making. The results show that the SARIMA model is the most accurate in predicting renewable energy consumption. This model has a Root Mean Square Error (RMSE) value of 0.0484 and a Mean Absolute Error (MAE) of 0.0196, and an R-Squared value of 0.6918. The study concludes that SARIMA is the most effective AI model for forecasting renewable energy consumption, significantly contributing to energy planning and sustainable management practices. This research provides a robust framework for managing the unpredictability of renewable energy data, supporting more informed energy policy and management decisions.