Research¶
Research Experience¶
Research Intern, Georgia Institute of Technology
Supervisor: Prof. Qi Tang, 2025.12-2026.04
- Collaborated weekly on structure-preserving operator learning for problems in plasma and fusion modeling.
- Designed a solution embedding a semi-Lagrangian discretization into a Fourier Neural Operator to bridge advection and diffusion-dominated regimes.
- Developed and validated a single model for both sharp shock structures and smooth convection-diffusion behavior.
Research Intern, Southern University of Science and Technology
Supervisor: Prof. Kailiang Wu, 2025.07-2025.08
- Translated an operator-splitting and machine-learning correction framework into a computational model for model-reality gaps.
- Studied stability and approximation properties of the proposed operator-learning framework.
- Implemented and benchmarked the framework on Navier-Stokes, FitzHugh-Nagumo, and one-dimensional compressible Euler systems.
Research Projects¶
Zhejiang Province SRTP Project: Application and Optimization of Classical Computational Methods and Machine Learning Techniques in Flow Field Simulation
Supervisor: Prof. Heyu Wang, 2024.03-2025.05
- Worked with a finite-element program for solving the Navier-Stokes equations and gained hands-on experience in computational fluid dynamics.
- Explored the integration of classical computational methods with machine learning to improve simulation accuracy and reduce computational cost.
- Investigated optimization techniques and loss functions for Physics-Informed Neural Network models.