Forecasting Indian Conventional Generation: A Comparative Analysis of ARIMA and SARIMA Models for Improved Energy Demand Prediction

This research presents an efficient approach to anticipate future conventional electricity generation, offering valuable insights into addressing upcoming energy demands. To unified models are introduced, Autoregressive Integrated Moving Average (ARIMA) and Seasonal Autoregressive Integrated Moving Average (SARIMA), tailored for predicting India's conventional electricity generation. This model is trained over a 40-month dataset and is sourced from the Central Electricity Authority (CEA), which utilizes autocorrelation and partial correlation plots to determine the optimal model order. Among these models, SARIMAX demonstrates superior accuracy over ARIMA, selected based on lower values of Root Mean Square Error (RMSE) and Mean Absolute Error (MAE) values. Using Machine Learning Algorithms, the current conventional trends can be analyzed, and effective methods can be employed to sustain stability in the power system grids in India. The models provide valuable insights into forthcoming load patterns and projected electricity demands by the year 2031.