Outsmarting the energy grid with AI
summary
Ixor’s cloud service ingests smart‑meter, tariff and weather feeds via EventBridge into AWS Lambda workers, stores history in DynamoDB (with TTL) and S3, and produces per‑home hourly forecasts using LightGBM models trained at inference time. A constraint‑optimization scheduler enforces capacity tariffs, user deadlines and dynamic pricing to output executable charging and battery plans. The pipeline targets linear per‑user cost and reports about €1.2 / month per user for 50‑minute retrain/inference cycles.
AI-written summary · about
chapters (9)
- 0:00Introduction and context
- 4:20Energy management complexity
- 9:39Live demo of the solution
- 12:32Cloud architecture overview
- 17:32Data handling and GDPR
- 18:38Forecasting and optimization
- 25:48User guidance and UX challenges
- 29:02Cost efficiency and scalability
- 30:09Q&A and conclusion
related talks
46mDigitizing energy management for a sustainable future with Iberdrola
18mFrom Smart Homes to Smart Grids
18mOptimizing energy operations with generative AI: TC Energy and Pariveda
20mFrom Hype to Value: How AI is transforming the utilities industry
28mCloud Computing for the Power Grid: From Service Composition to Assured Clouds
44mSmart application on Azure at Vattenfall
40mDeep Learning and Time Series Forecasting for Smarter Energy
21mBest Practices for Optimizing Your Cloud Spend with AI