2026
Multiscale Hypersonic Boundary Layer Reconstruction via Spectral Binning and Subdomain-wise Conditional Diffusion
6Publications
12+Talks & posters
3Journals reviewed
SNU→PSU→PurduePath
About
I am a Ph.D. student in Mechanical Engineering at Purdue University, advised by Prof. Romit Maulik. My work sits at the intersection of computational fluid dynamics and machine learning: I develop scientific machine learning methods that are generalizable, interpretable, and consistent with the underlying physics.
I am particularly interested in differentiable-physics approaches to turbulence closure modeling on unstructured grids, interpretable deep learning for dynamical systems, and generative models for reconstructing high-speed flows. Previously, I earned my B.S. and M.S. in Aerospace Engineering from Seoul National University. In summer 2026, I will join Lawrence Livermore National Laboratory as a Graduate Research Intern.
News
- May 2026Starting a summer research internship at Lawrence Livermore National Laboratory on ML for multiphase fluid dynamics.
- Jun 2026Presenting our differentiable-physics turbulence closure work at USNC-TAM, Pasadena.
- Mar 2026New preprint on conditional diffusion for hypersonic boundary-layer reconstruction (arXiv:2606.15023).
- Jan 2026Talk on interpreting learned physics in Neural ODEs via DEIM at the Joint Mathematics Meetings.
Publications
H. Kim denotes the author of this site.
2025
Towards Interpretable Deep Learning and Analysis of Dynamical Systems via the Discrete Empirical Interpolation Method
2026
Generalizable data-driven turbulence closure modeling on unstructured grids with differentiable physics
2025
Interpretable A-posteriori error indication for graph neural network surrogate models
2025
Improved Conceptual Design of eVTOL Aircraft: Considering Rotor–Rotor Interactional Effects
2024
Data-driven physics-informed neural networks: A digital twin perspective
Selected Talks & Conferences
- Generalizable data-driven turbulence closure modeling on unstructured grids with differentiable physicsUSNC-TAM · Pasadena, CA · Jun 2026
- Conditional Diffusion Models for High-speed Boundary Layer ReconstructionsSIAM UQ26 · Minneapolis, MN · Mar 2026
- Interpreting Learned Physics in Neural ODEs through a DEIM-based Spatiotemporal AnalysisJoint Mathematics Meetings · Washington DC · Jan 2026
- Interpreting Learned Physics in Neural ODEs through a DEIM-based Spatiotemporal AnalysisAPS DFD 78th Meeting · Houston, TX · Nov 2025
- Differentiable physics for generalizable closure modeling of separated flowsSIAM CSE25 · Fort Worth, TX · Mar 2025
- Differentiable physics for generalizable closure modeling of separated flowsAPS DFD 77th Meeting · Salt Lake City, UT · Nov 2024
Workshops & earlier presentations
- Towards Interpretable Deep Learning via the Discrete Empirical Interpolation MethodAAAI XAI4Science Workshop · Singapore · Jan 2026
- Scalable, interpretable, and explainable scientific machine learning with geometric deep learning18th USNCCM · Chicago, IL · Jul 2025
- Differentiable physics for generalizable closure modeling of separated flowsPurdue CCAM SciML Workshop · West Lafayette, IN · Sep 2025
- Scalable, adaptive, and explainable scientific machine learning for surrogate models of PDEsAAAI Bridge Program (KGML) · Philadelphia, PA · Feb 2025
- Scalable, adaptive, and explainable scientific machine learning for surrogate models of PDEsAAAI Symposium on Computational Scientific Discovery · Arlington, VA · Nov 2024
- Design Methodology of Urban Air Mobility for Noise Mitigation at the Conceptual Design Stage48th European Rotorcraft Forum · Winterthur, Switzerland · Sep 2022
Honors & Awards
- 2026Purdue Graduate Student Government Travel Award
- 2025Selected participant, Argonne Training Program on Extreme-Scale Computing (ATPESC)
- 2025Student Travel Award, SIAM CSE25
- 2024Ruth Young Boucke Graduate Fellowship, Penn State
- 2024Robert W. Graham Endowed Graduate Fellowship, Penn State
- 2015–22Merit-based & Work-study Scholarships, Seoul National University
Professional Service
Journal reviewer for Journal of Fluid Mechanics, Journal of Computational Physics, and Physics of Fluids.