Electricity Demand Forecasting in the Northern Region of Bangladesh: A Deep Learning Approach

This study addresses the challenge of accurately forecasting electricity demand in the Northern Region of Bangladesh, a task critical for optimizing power generation. A deep learning approach combined CNN-GRU, DLSTM, and N-BEATS models, with predictions aggregated through a weighted average. The proposed models demonstrated significantly lower mean absolute errors with most consumers achieving accuracy above 98 %, indicating their potential for providing more reliable demand forecasts. These findings suggest that the adoption of these advanced models could improve decision-making processes in electricity generation, leading to a more efficient power grid.