Conceptual visualization
Research area 01
CFD + AI
Physics-aware surrogate models, turbulence modelling, flow reconstruction, inverse design, and uncertainty-aware optimization.
Explore CFD + AIResearch group
Artificial Intelligence
and Simulations
Research Group
Research at the intersection of artificial intelligence and computational simulation.
Our current focus connects physics-aware learning with fluid flow and phase-change thermal systems.
Focus areas
Two connected research tracks.
We use simulation data, physical constraints, and careful validation to study faster predictive models and better engineering decisions.
Conceptual visualization
Conceptual visualization
Research area 02
PCM + AI
Data-guided phase-change materials, melt-front prediction, thermal-storage design, and intelligent charge/discharge control.
Explore PCM + AIResearch loop
Simulation → Learning → Validation
- 01Formulate
Define the physical problem, assumptions, operating envelope, and quantities of interest.
- 02Simulate
Generate traceable field and integral data with numerical models and, where available, experiments.
- 03Learn
Build models that use structure from equations, boundary conditions, symmetries, and data.
- 04Validate
Test conservation, stability, uncertainty, and performance beyond the training cases.
Research principle: AI predictions are engineering evidence only after comparison with trusted numerical or experimental references.