Tianjia Huang

Tianjia Huang

Postdoctoral Associate | Subsurface Energy Systems Modeling | High-Performance Scientific Computing

Massachusetts Institute of Technology

Email: huangtj@mit.edu

CV: Download


Education


Research Interests

I am a Postdoctoral Associate at MIT working on subsurface energy systems, reactive transport, and computational geomechanics. My research combines high-fidelity numerical modeling with high-performance scientific computing to study coupled thermal-hydraulic-mechanical-chemical processes in geologic media. A central theme of my work is extending the TOUGH family of simulators so that fundamental model development translates into tools that can be used for underground hydrogen storage, carbon storage, geothermal energy, natural gas hydrates, and nuclear waste disposal.

  1. Coupled thermal-hydraulic-mechanical processes in subsurface systems
  2. Fracture network representations, such as DFN and EDFN
  3. Particle transport and solid-fluid interactions in porous media
  4. High-performance scientific computing for energy and environmental applications

Research Impact & Scientific Software Development

  • Parallel TOUGH+RealGasBrine Simulator (Fortran & MPI)
    Lead Developer: Tianjia Huang & George Moridis
    Code Access: Serial code; parallel code will be released soon.
    • Developed distributed-memory MPI execution, PETSc solver integration, and dynamic domain repartitioning for non-isothermal compositional multiphase flow.
    • Reduced runtime from 1,360 to 20 minutes on 64 MPI ranks for a benchmark.
    • Accelerated the serial version with OpenMP for EOS evaluation and Jacobian assembly and a GPU solver for matrix solving.
    • Coupled the flow simulator with an in-house parallel geomechanics code for distributed-memory THM simulation.
    • Applied the simulator to underground hydrogen storage in aquifers and depleted hydrocarbon reservoirs, carbon sequestration in saline formations, and other coupled-process subsurface energy studies.
  • Parallel Particle Transport Module for TOUGH+Hydrate (Fortran & MPI)
    Lead Developer: Tianjia Huang & George Moridis
    Code Access: TOUGH+Hydrate; the standalone particle transport module will be open-sourced on my GitHub.
    • Developed a discrete element method module coupled with TOUGH+Hydrate to simulate particle detachment, migration, and clogging in hydrate-bearing sediments.
    • Implemented reservoir-scale simulations involving millions of particles to investigate sand production during gas extraction.
  • Multiphase Flash Calculation Algorithms (MATLAB & Python)
    Lead Developer: Tianjia Huang & Russell Johns
    Code Access: Research group products page or my GitHub.
    • Developed robust algorithms for solving multiphase Rachford-Rice equations with arbitrary numbers of phases and components.
  • Geothermal Reactive Transport Modeling Framework (Python)
    Lead Developer: Tianjia Huang
    Code Access: my GitHub.
    • Developed a Python-based framework integrating multiple open-source mesh generation tools to construct three-dimensional matrix-fracture systems for finite-volume reactive transport simulations.
    • Coupled reactive transport models with discrete fracture network representations to analyze mineral precipitation, permeability evolution, and injectivity decline in geothermal reservoirs.
  • Computational Modeling for Subsurface Energy Systems (TOUGH)
    Lead Developer: Tianjia Huang
    Code Access: Please feel free to contact me for collaboration.
    Simulation frameworks developed in this research support modeling of:
    • Underground hydrogen storage and geological hydrogen
    • Geothermal reservoir evolution
    • Nuclear waste repository performance
    • Natural gas hydrate production

Work Experience