Long Term Load Forecasting Model for Chhattisgarh State

Long term load forecasting (LTLF) is a crucial task in power system planning, and the selection of appropriate forecasting techniques is essential to ensure accurate predictions. In this study, three popular techniques, namely Long Short-Term Memory (LSTM), Artificial Neural Network (ANN), and Gaussian Process Regression (GPR), are utilized for LTLF in Chhattisgarh, India. The input factors considered for the forecasting models include generation and load demand data for the years 2020 and 2021. The models are trained using historical generation and load data for specified time period, and their performance is evaluated using statistical metrics such as mean absolute error (MAE), root mean square error (RMSE), and mean absolute percentage error (MAPE). The result obtained demonstrate that the LSTM model outperforms both ANN and GPR model in terms of forecasting accuracy for specified region and time period. The LSTM models showcases superior predictive capability, providing the most accurate load forecasts compared to other models.