Lightning Talk: Forecasted Congestion Management Using Plug and Play Model
summary
Builds a production plug and play model that ingests messy CGMES/UCTE exports from SCADA, runs powsybl, pandapower, PGM and VeraGrid via adapter bridges, and orchestrates containers on Kubernetes with Kafka triggers. An AI Monte Carlo classifier picks the best importer and solver and configures self-healing fallback chains; the platform serves >50 grid operators and handles grids representing ~50% of installed wind/solar forecasted capacity. Real cases: a West German outage only converges with a PGM solver; a Bavarian state estimation is debugged with VeraGrid, showing the approach works in production.
AI-written summary · about
chapters (7)
- 0:00Introduction and background
- 1:47Business case and challenges
- 4:57Solution approach and innovation
- 9:09AI-driven fallback strategy
- 12:03Real-world examples and results
- 13:27Call for collaboration and open source
- 14:55Q&A session
related talks
14mLF Energy’s Power Grid Model Project Tackles Complex Grid Calculations Efficiently
29mTaming the Circus: The Future of Power System Planning Model Interoperability
31mRedispatch Without Black Boxes: Load Forecasting With OpenSTEF at E.ON
32mCIM/CGMES in Real-time Knowledge Graphs for Modern Grid Control Systems
33mFrom Open Data To Foundation Models: Building GridSFM for Power Grid An...
15mMaking Complex R Forecast App...
13mOpen Source Can Help With How We Consume And Produce Electricity
06mLightning Talk: Leveraging Realtime Geospatial Data and the Spring Ecosystem to Save Lives, Solstice