Investigation of 1D system CFD and 3D CFD numerical methodology applied to an experimental facility

Abstract

The fourth generation gas-cooled nuclear reactor designs provide a promising prospect for energy and process heat generation. The designs` cycles have high thermal efficiencies, require relatively little fuel and are safe to operate. The process of fission does not emit carbon dioxide therefore not adding more greenhouse gases to the atmosphere. Nuclear power stations are subject to strict regulations and safety standards since an accident could have severe consequences. The regulations stipulate, amongst other things, that the nuclear system`s behaviour must be able to be predicted under all conditions. To this end, the computational methods used to predict the behaviour must be verified and validated. The phenomena in a nuclear reactor can be very complex and computationally expensive to model, especially when using a 3D CFD approach. The use of 1D system CFD can be employed with improvements in computational time, but with limitations in terms of detail. 1D methodology must therefore be verified and validated against 3D methodologies and experiments to ensure that all of the relevant phenomena are accounted for. The reactor cavity cooling system experimental facility at the University of Wisconsin was simulated by using a combination of 1D methodology by using Flownex SE and 3D methodologies by using ANSYS Fluent. The facility is a scale model of the reactor cavity cooling system (RCCS) of a modular high temperature gas cooled reactor (MHTGR). The RCCS operates solely on buoyancy forces, making it independent of both operator input and power source. The buoyancy driven flow also required the proper correlations be used when numerically simulating the phenomena. The coolant loop consists of a pipe network, which is fed from a tank, that passes through a heated cavity (the latter emulates the reactor cavity). Various levels of heat were added during experiments at the heated cavity, simulating various conditions that could occur in a full scale prototype. The experimental conditions were used as boundary conditions in the CFD/system CFD simulations and the results were compared. The heated cavity and the water tank necessitated the use of 3D CFD methodologies, while a 1D approach was used in the other parts of the system. The numerical results obtained by simulation compare well with the experimental results