New ensemble learning model forecasts electricity use 12 months ahead

Researchers have developed a cooperative ensemble learning approach called Weaker Separator Booster (WSB) to forecast electricity consumption up to 12 months in advance. The study utilized historical data from two campuses of the Federal Institute of Paraná (IFPR) and incorporated feature selection using SHAP and optimization via GA and PSO. The WSB model demonstrated superior performance, achieving an sMAPE of 13.90% and MAE of 1990.87 kWh for one campus, and 18.72% sMAPE and 465.02 kWh MAE for the other. Analysis indicated that lagged time-series values were the most influential factors, while climatic variables had minimal impact. AI

IMPACT This research presents a novel ensemble learning method for improving long-term electricity consumption forecasting accuracy.

RANK_REASON The cluster contains an academic paper detailing a new machine learning approach for forecasting. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.LG →

• extreme gradient boosting (XGBoost)
• Federal Institute of Paraná (IFPR)
• LSTM
• Weaker Separator Booster (WSB)

• paper
• other

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →