The Value of Open Source Short-term Energy Forecasting Demo...
summary
openSTEF is a Python framework that delivers probabilistic forecasts hours to days ahead and runs end-to-end pipelines (data prep, feature engineering, post‑processing, evaluation). It powers congestion management at a large Dutch DSO: per‑asset and MV‑route forecasts trigger flexible contracts to avoid overloads, nowcasting fills 1–2 day measurement delays for real-time nowcasting, and transport and grid‑loss forecasts feed day‑ahead purchases. A district‑heating operator uses openSTEF inside Aurora to optimize boilers and a thermal battery, and multiple per‑site models (including XGBoost) run in production with reported national‑scale adoption.
AI-written summary · about
chapters (15)
- 0:00Introduction and speakers
- 1:25Grid congestion challenges
- 3:08Grid congestion forecasting
- 4:33Medium voltage route monitoring
- 6:58Free space forecasting
- 8:46Transport forecast for TSO
- 10:32Grid losses forecasting
- 12:41Summary of Alander use cases
- 13:02District heating overview
- 15:06Heat production optimization
- 19:25OpenStaff for heat forecast
- 20:48Adopting OpenStaff experience
- 22:13Impact in Swedish district heating
- 23:16Model selection for forecasts
- 24:00Key takeaways and contact
related talks
11mOpenSTEF: Forecasting The Load On An Energy Grid
27mOpenSTEF 4 Launch: A New Way of Building Energy...
32mAdopting Open-Source Tools for Time Series Forecasting
27mOpen Source in an Age of Transformation for the Energy Industry
32mPredico: A Collaborative Forecasting Market for the Energy Sector
13mHow Open Source Is Helping Fight Climate Change
13mOpen Source Can Help With How We Consume And Produce Electricity
33mFrom Open Data To Foundation Models: Building GridSFM for Power Grid An...