Machine Learning and Time-Series Foundation Models for Day-Ahead Electricity Forecasting in a University Building

Key points

• Day-ahead electricity forecasts can help building operators schedule heating, ventilation, and other services around anticipated demand.1
• Choosing a forecasting model also requires balancing accuracy with its forecast-time inputs, local training, and computational requirements.1
• An exploratory historical evaluation reveals seasonal variation, while a separate weather-source comparison produces only small changes in the selected TimesFM configuration's accuracy.1
• These results support pretrained TimesFM as a strong initial forecasting baseline for this building, with XGBoost as a competitive locally fitted alternative.1

Sources (1)

• [1]Machine Learning and Time-Series Foundation Models for Day-Ahead Electricity Forecasting in a University Building
  arXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 8, 07:00 PM

  We benchmark statistical, machine-learning, recurrent neural-network, and pretrained TimesFM models for a university building.

  Day-ahead electricity forecasts can help building operators schedule heating, ventilation, and other services around anticipated demand.

Extractive summary: sentences quoted from the sources.