Direct numerical simulation of two-phase stratified flows in the primary coolant of a nuclear reactor

Abstract

Multiphase flows are very common in many Nuclear engineering applications. During high pressurized conditions there are possibilities of high thermal loads on the pressure vessel, leading to pipe ruptures. As part of breakdown measures, the emergency core cooling system is activated and the coolant is mixed with the fluid in the cold leg, giving rise to multiphase turbulent flow. These regimes can comprise of large scale interfaces, leading to stratified flows. These postulated accidents or events need to be identified and understood to improve nuclear reactor safety. Computational fluid dynamics can serve as an excellent tool to model these scenarios, contributes towards reactor safety. Coarse models which are widely used in industries such as RANS are known to over-predict turbulent producing unphysical gradients. Thus the turbulent mass and momentum are not yet fully understood. Using high resolution tools such as Direct Numerical Simulations (DNS), can potentially avoid these over-prediction and could model these large scale interfaces accurately. As a long term goal, the data sets generated from these simulations can be used to train such coarse models or simply support for validation.

Placing the focus on a configuration where two fluids are in a stratified scenario, this graduation thesis will show a systematic approach towards the development and modelling of air and water moving in both co-current and counter-current direction, wherein simulations are performed in RK-Basilisk. Primarily, the work starts with studying a single phase turbulent channel flow to forma basis of understanding of concepts and code. The model of (Liu et al. 2009), who use realistic properties of air-water is chosen to be implemented in RK-Basilisk. It is realized that, implementing this is in RK-Basilisk is not straightforward and thus the constraints are identified and a general mathematical framework is developed to resolve this. One of the main objectives in this thesis is to model and understand the turbulent behavior near the interface of both air and water. To do so, the physical mechanisms which govern the generation and decay of turbulence called the TKE Budgets is studied by modelling the individual terms that complete it. The budgets are modelled and validated against (Liu et al. 2009). Interesting conclusions are drawn which depict the trends of budget terms and the kinetic energy, giving a good picture of the underlying interfacial turbulent mechanisms. The same mathematical framework, along with some additional modelling lead to an extension of this study to counter-current flows, wherein another set of
conclusions are drawn.