The Physics-based Digital Transformation Lab aims to devise and harness rapid yet accurate simulation technology for the entire life cycle of engineering systems. For instance, we can utilize such simulations not only for design during product development but also for condition-based monitoring and maintenance during product operation. Our research applications encompass various areas, such as product design, prognostics and health management, digital twins, and more. Our research techniques include model order reduction, particularly reduced basis methods, as well as machine learning and AI algorithms.

1. Machine learning assisted a posteriori error estimation of reduced basis approximation
2. Lego-like composite material construction and analysis on the fly
3. Rapid yet accurate component-based wing structure simulation for design and digital twin applications
4. Multi-scale reduced basis homogenization for rapid yet accurate material property prediction
5. Reliability-based design optimization using reduced basis approximation
6. Condition-based monitoring and maintenance using a reduced basis digital twin
7. System identification of an unmaned underwater vehicle by AI technology
8. PIV data restoration based on a low-dimensional flow structure
9. Multi-fidelity reduced order modeling with manifold alignment for aerodynamic simulation