Nuclear reactors need to be designed to ensure materials aren’t overheated, and historically, researchers used crude geometry models and expensive mockup experiments. Being conservative was the name of the game.
Now, a team from the Department of Energy (DoE) devised a high-fidelity approach where they can assess the heat distribution in a nuclear reactor design with fewer or no modeling approximations.
This approach shows high-fidelity modeling capabilities can improve the economics and safety of advanced fast reactor designs. Specifically, reactors can be designed and operated safely at higher power levels, which can lead economic gains.
This work supports the DoE’s mission where they can facilitate commercial deployment of reactors as a clean and sustainable energy source.
In testing, researchers estimated hot channel factors (HCFs) with the high-fidelity Simulation-based High-efficiency Advanced Reactor Prototyping (SHARP) toolkit by modeling pin-by-pin nuclear fuel assemblies subjected to various uncertainties (wire wrap vs. bare bundle, fuel fissile content maldistribution, cladding thickness manufacturing uncertainty, and uncertainties in thermo-physical properties).
As part of this research, they used the Mira supercomputer at the Argonne Leadership Computing Facility, a DoE Office of Science user facility.
With data from Mira, the team compared the new HCFs to legacy HCFs evaluated for a sodium-cooled fast reactor, EBR-II, and showed reductions in every case.
The legacy HCFs have built-in conservatism due to their crude geometry models and approximate physics modeling.
By removing this conservatism with high-fidelity models, engineers can now design and operate reactors safely at higher power levels, leading to economic gains.