IEEE International Conference (QPAIN 2026)
2026
Advanced Hybrid Deep Learning Models for Solar Energy Holding Capacity Prediction in Bangladesh
Bangladesh receives abundant solar energy throughout the year, yet actual generation often falls short due to weather variability. This study proposes a hybrid stacking ensemble model combining Random Forest, XGBoost, Gradient Boosting, and AdaBoost with an MLP meta-learner to predict solar energy holding capacity. The model achieves up to 95% accuracy and outperforms individual models, supporting efficient renewable energy planning and policy-making.
0 citations Altmetric 0 deep-learning · ensemble-learning · solar-energy · energy-forecasting · bangladesh