Dr. Amirmehran Mahdavi works in the field of rarefied gas dynamics and computational modeling, with a focus on developing advanced numerical frameworks for microscale and non-equilibrium flows. His research spans Direct Simulation Monte Carlo (DSMC) methods, Fokker–Planck–based kinetic modeling, and innovative hybrid DSMC–FP algorithms that significantly improve efficiency while preserving molecular-level accuracy.
He has also explored complex thermal and flow phenomena in micro/nano geometries, including thermal separation, cold-to-hot energy transport, and rarefaction-induced flow behavior. More recently, he has integrated machine learning, particularly Deep Operator Networks (DeepONet), to develop fast and physics-informed surrogate models capable of reproducing DSMC results with several orders-of-magnitude speedup.
In addition to rarefied gas dynamics, Dr. Mahdavi has contributed to the study of cavitation and flow instabilities in turbomachinery and hydrodynamic systems, investigating strategies to mitigate destructive effects using advanced modeling and bio-inspired concepts. His research collectively aims to bridge first-principles physics, computational efficiency, and intelligent data-driven modeling for next-generation engineering applications.