Digital Battery Twin
summary
Company stack combines SAP Asset Management, TWAICE predictive analytics and MHP integration to build per-vehicle digital battery twins that use fleet telemetry and lab simulators to deliver real-time health and lifetime estimates. Hybrid modeling mixes surrogate models that learn simulation input→output for tiny edge predictors, physics-informed neural networks to embed conservation laws, and corrective models that patch slow or brittle physical simulators. Outputs power predictive maintenance, per-car charging schedules and warranty decisions; prototypes are live in OEM pilots and update continuously from fleet data.
AI-written summary · about
chapters (12)
- 0:00Intro
- 0:32SAP Asset Management Solutions (https://youtu.be/aXayGGzsSKQ)
- 3:30Talk intro
- 5:45Digital twins
- 7:44Enabling technologies
- 9:22Machine learning
- 11:28The digital twin
- 13:11Continuous learning & understanding
- 15:35Hybrid modeling approaches
- 19:02Advantages of digital twins
- 19:42Use case: Digital battery twins
- 23:26Outro
related talks
01h 23mReal-Time Structural Health Monitoring Using Physics-Driven Digital Twins
27mGeneral Keynote Multi-tiered digital twins with what-if analysis...
26mData Driven Culture in Software Defined Vehicles
24mCircular Economy & Data: The New Battery Ecosystem
10mAI-Powered Digital Twins: Patterns for Smart Manufacturing
59mInnovating power plant operations through digital twins
20mEnvironmental Benefits of Efficient Power Moding in Vehicles
26mBridging the Gap: Embedded Deployment of PyTorch Deep Learning Models in Complex Engineered Systems