The purposes of containment spray system operation during a severe accident in a light water reactor (LWR) nuclear power plant (NPP) are to depressurize the containment by steam condensation on spray droplets, to reduce the risk of hydrogen burning by mixing the containment atmosphere, and to collect radioactive aerosols from the containment atmosphere. While the depressurization may be predicted fairly well using lumped-parameter codes, the prediction of mixing and collection of aerosols requires a local description of transport phenomena. In the present work, modelling of sprays on local instantenous scale is presented and the Design of Experiment (DOE) method is used to assess the influence of boundary conditions on the simulation results. The TOSQAN 101 spray test, which was used for a benchmarking exercise within the EU Severe accident research network of excellence (SARNET), was simulated, and simulation results were compared to experimental data. The modelling approach is based on a Lagrangian description of the dispersed liquid phase (droplets), an Eulerian approach for the description of the continuous gas phase, and a two-way interaction between the phases. The simulations are performed using a combination of the computational fluid dynamics (CFD) code CFX4.4, which solves the gas transport equations, and of a newly proposed dedicated Lagrangian droplet-tracking code. The intent of the presented work is to assess the modeling of sprays and liquid on the wall in the presented approach with the emphasis on the heat and mass transfer between liquid and gas phase. The simulation-experiment comparison of available global and local variables demonstrates that the proposed approach is suitable for prediction of global variables evolution and of the non-homogenous structure of the atmosphere. The boundary condition steady-state sensitivity study is performed and shows that the global variables are mostly affected by the wall temperature boundary condition.

This content is only available via PDF.
You do not currently have access to this content.