Computational Modeling of Boiling Flow Regime Maps
Bernard Geurts & Hans Kuerten
Turbulent multiphase flow is a subject of intense ongoing research. Numerical simulation represents a powerful tool for analysing and understanding the complex flow structures that typically arise in such physical problems. The so called ‘direct numerical simulation’ (DNS) allows for full resolution up to the smallest scale of motion producing a database of physical relevance. The presence of the second phase enriches the problem making it numerically more challenging. In this project, we focus on turbulent bubble-laden channel flow investigating changes in the flow features and in the transport of heat as a function of the bubble volume fraction.
The intended main results is a multi-scale numerical method to determine flow pattern maps, validated using experimental data at operation conditions. To this end we developed an in-house numerical code designed to solve efficiently and accurately this specific problem. The proposed research will lay the groundwork for the analysis of practical flow pattern maps for heated two-phase flows.
-developer: Paolo Cifani
- Third/Second order finite-differences solver
- Navier-Stokes equations on staggered cartesian mesh
- Wall bounded turbulent flow
- Volume of Fluid (VOF) tracking on second phase
- Bubble-Bubble interaction
- On-the-fly computations of flow statistics
- MPI-OpenMP hybrid parallelisation
Parallel scalability of the solver mphBox for a turbulent channel flow on a 33 million grid. The total gas volume fraction is 10% corresponding to about 1000 bubbles. The black line is total wall-time per time step; the red line is wall-time of the Poisson solver; the blue line is the wall-time of the volume fraction solver. Linear scaling is shown up to 400 cores.
Stream-wise velocity field (left figure) and span-wise vorticity field (right figure) on a mid-plane of a turbulent channel flow. Significant amount of small structures in the core of the channel are induced by the passing of the bubbles.