Xiaoxuan Shao

Research Student at Institute of Information Management in Mechanical Engineering (IMA), RWTH Aachen University

Combined physics-based CFD simulations with data-driven learning to study how structured physical priors can improve learning efficiency and generalization; investigated hybrid modeling strategies that reduce computational complexity by transferring knowledge from high-fidelity simulations to lightweight learning models; demonstrated that incorporating domain structure into learning systems enables compact models suitable for near-real-time and resource-constrained deployment.