Using supercomputers, researchers will simulate how hot helium mixes with air during rare cooling emergencies in advanced reactors, producing data that helps engineers design safer systems and prevent oxygen-related risks.
Advanced reactors such as High Temperature Gas Reactors (HTGR), are being developed by U.S. companies for deployment in the late 2020s or early 2030s. Their development requires extensive computational work involving accurate fluid dynamic simulations under normal operation and accident scenarios. One of the critical safety design tests for HTGRs is the ability to dissipate decay heat safely during Depressurized Conduction Cooling (DCC) conditions. In one accident scenario of HTGR operation, hot helium gas will be discharged from a high-pressure pipe into a cavity surrounding a Reactor Pressure Vessel (RPV) or steam generator. Hot helium jets will mix with the cold air in the cavity and this gas mixture may be vented through different locations in the reactor. The amount of air entering the reactor is significantly governed by helium-air mixing. Transmission of the gas mixture is a major concern as oxygen migration leads to the risk of exothermic (combustible) reactions. Therefore assessing the concentration levels of the species (helium, oxygen, nitrogen, etc.) become important to understanding the impacts of the accidental scenario.
The simulations in this project are a continuation of the work in previous ALCC 2024/25 project which used Nek5000 and NekRS for simulating helium-air mixing in HTGR cavities. This project will utilize the Polaris supercomputer to gather high resolution velocity, temperature and gas-concentration data and will provide a critical dataset to compare against the experiments already performed or currently underway at CCNY. The objective will be to build the dataset by including two additional design components to the simulations: (1) multiple cavities connected via vents and (2) variable location of the helium-jet entrance. Simulations that include both of these design features will help augment the dataset to help validate codes and improve our understanding of the heat transfer processes in HTGRs.