Dr. Jun Fang
Nuclear Engineer
Argonne National Laboratory

Abstract

Complex flow structure interactions and heat transfer processes take place in nuclear reactor cores. Given the extreme pressure/temperature and radioactive conditions inside the core, numerical simulations offer an attractive and sometimes more feasible approach to study the related flow and heat transfer phenomena in addition to the experiments. Under the Exascale Computing Project, the full-core simulation of a small modular reactor (SMR) has been pursued coupling Computational Fluid Dynamics (CFD) and neutronics. A key aspect of the modeling of SMR fuel assemblies is the presence of spacer grids and the mixing promoted by mixing vanes or the equivalent. A reduced order methodology is adopted based on momentum sources to mimic the mixing of the vanes. The momentum sources have been carefully calibrated with detailed Large Eddy Simulations (LES) of spacer grids performed with Nek5000. Modeling the spacer grid and mixing vanes (SGMV) effect without body-fitted computational grid avoids the excessive costs in resolving the local geometric details, and thus supports the simulation to be scaled up to the full core. Besides the progress on momentum source modeling, this study also features the first full-core pin resolved CFD simulation ever performed to the author’s knowledge. This represents a significant advancement in capability for the CFD of nuclear reactors, which will hopefully serve as an inspiration for further integrating high-fidelity numerical simulations in actual engineering designs.

Biography

Dr. Jun Fang is currently a nuclear engineer in the Nuclear Science and Engineering Division at the Argonne National Laboratory. His research interests include reactor thermal hydraulics analyses and two-phase flow modeling leveraging cutting-edge simulation techniques and high-performance computing. Dr. Fang has 9 years of experience in high-fidelity simulations of both Light Water Reactors (LWR) and non-LWR advanced reactors on leadership class supercomputers. One of his current projects is focused on developing a coupled multiphysics model of the entire Small Modular Reactor core for the upcoming Exascale computing platforms coupling both CFD and neutronics. He received his Ph.D. degree in nuclear engineering from North Carolina State University in 2016.

Thursday, October 21. 2021
4:00 pm seminar

Hybrid Option (Speaker is remote)

Zoom (link upon request)
or
Room 1202 Burlington Labs
(Refreshments)