About
My research advances scientific machine learning for multi-physics simulation, combining physics-constrained neural networks with data-driven sub-grid-scale closure modeling to capture spatiotemporal atmospheric boundary layer dynamics. I develop invariance-embedded machine learning surrogates and integrate them into large-scale CFD solvers such as WRF, using high-performance computing to run high-fidelity hurricane boundary layer simulations at scale.
I completed my PhD in Mechanical Engineering at the University of Maryland, Baltimore County in June 2026, advised by Dr. Meilin Yu. My dissertation covered model development, a priori assessment, and a posteriori tests of these closures inside WRF.
I also work with radar and remote sensing datasets, high-performance GPU computing, and physics-informed neural networks for environmental systems. My broader interests include hurricane hazard modeling, climate resilience, and integrating machine learning with large-scale numerical prediction systems.
Selected publications
-
2026
Invariance-embedded machine learning sub-grid-scale stress models for mesoscale hurricane boundary layer flow simulation I: model development and a priori studies
International Journal for Numerical Methods in Fluids
Accepted arXiv
-
2025
Comparison of several machine-learning-enhanced sub-grid scale stress models for mesoscale hurricane boundary layer flow simulation
AIAA SciTech 2025 Forum, p. 2212
Best student paper DOI
-
2022
The effects of numerical dissipation on hurricane rapid intensification with observational heating
Journal of Advances in Modeling Earth Systems, 14, e2021MS002897
Education
PhD, Mechanical Engineering
University of Maryland, Baltimore County. Advisor: Dr. Meilin Yu.
MS, Mechanical Engineering
University of Maryland, Baltimore County. Advisors: Dr. Meilin Yu, Dr. Stephen Guimond.
BSc, Mechanical Engineering
Bangladesh University of Engineering and Technology
Experience
Graduate Research Assistant
Computational Mechanics Laboratory, UMBC
- Conducted a priori and a posteriori evaluations of invariance-embedded machine learning sub-grid-scale closures for rotating hurricane boundary layers, integrating spatio-temporal stress and eddy-viscosity surrogates into WRF. Project supported by the UMBC 2025 COEIT Interdisciplinary Proposal Award.
- Developed machine-learning models of backscatter-admitting sub-grid-scale processes to improve hurricane boundary layer simulation.
- Advanced methods for detecting stealthy, long-term cyber-attacks on wind energy assets using physics-informed neural networks. Project supported by the UMBC 2024 Cybersecurity Leadership Exploratory Grant.
- Contributed to the development of next-generation naval design tools using CFDship-Iowa, with industrial and academic partners.
Graduate Research Assistant
Joint Center for Earth Systems Technology, UMBC
- Investigated numerical dissipation across weather prediction models (WRF and NUMA) with Dr. Stephen Guimond, supported by NSF grant AGS-2121366.
- Analyzed remote-sensing radar observations from the Imaging Wind and Rain Airborne Profiler (IWRAP) for the NOAA/AOML/HRD Hurricane Field Program.
Graduate Teaching Assistant
University of Maryland, Baltimore County
- Ran lab demonstrations and grading for ENME 432L, Fluids and Energy Lab, with Dr. Meilin Yu.
- Supported Dr. James Baughan with teaching and grading for ENME 423, HVAC Design.