Scale-up Technique of Slurry Pipelines—Part 2: Numerical Integration

1986 ◽  
Vol 108 (4) ◽  
pp. 278-285
Author(s):  
M. C. Roco ◽  
S. Mahadevan

A kinetic energy turbulence model has been proposed for the computer flow simulation and scale-up of slurry pipelines (in Part 1 [1]). The numerical integration is performed by using a modified finite volume technique, with application to high-convective two-phase flows in two and three dimensions (in Part 2). The mixture kinetic energy and eddy viscosity turbulence models are compared. The one-equation eddy-viscosity turbulence model (εt - model) is formulated in Part 2 and applied for the multi-species particle slurry flow in cylindrical pipes. A modified finite volume technique is proposed for high convective transport equations, for one and two-phase flows. The integral formulation per volume yields surface and volume integrals, that are stored and counted only by interfaces using a multidimensional approach. The nonlinear distributions in volumes and on interfaces are approximated employing the derivatives in the normal and tangent directions to the bounding surfaces. Linear, analytical (upwind) and logarithmic laws of interpolations are considered for internal flows. The numerical approach was tested with good results for transport equations of momentum and various contaminants (solid particles, temperature, eddy-viscosity) in pipes. Experimental data for one and two-phase flows are compared to the integral finite volume predictions. The proposed finite volume technique can economically simulate complex flow situations encountered in the slurry pipeline scale-up applications.

1986 ◽  
Vol 108 (4) ◽  
pp. 269-277 ◽  
Author(s):  
M. C. Roco ◽  
S. Mahadevan

A kinetic energy turbulence model is proposed for the flow simulation and scale-up of slurry pipelines (in Part 1). The numerical integration is performed by using a modified finite volume technique with application to high convective, two-phase flows, in two and three dimensions (in Part 2 [1]). The mixture kinetic energy and eddy viscosity one-equation turbulence models are compared. The constitutive equations and model constants are tested using laboratory experiments and then employed for large-scale applications. The governing equations are derived from the space/time averaging of the momentum equations and integrated in the pipe cross section using the finite volume approach. The specific interaction stresses (liquid-liquid, liquid-solid, solid-solid and solid-wall) are expressed in the mathematical formulation. The predictions for the velocity and concentration distributions, as well as on the mean velocity-headloss correlations, have been compared to available experimental data (water-sand, water-glass, water-coal mixtures; of concentrations αS = 5 – 40 vol percent, in pipes of various diameters D = 40 – 500 mm). The suggested model can simulate multi-species particulate pipe flow for which the semiempirical methods cannot be satisfactorily applied. The numerical tests and comparison to experiments show the model capabilities to scale-up data from laboratory to real flow situations via infinitesimal two-phase flow analysis.


2020 ◽  
Vol 213 ◽  
pp. 104715
Author(s):  
Rihua Yang ◽  
Heng Li ◽  
Aiming Yang

1983 ◽  
Vol 26 (4) ◽  
pp. 931 ◽  
Author(s):  
S. E. Elghobashi

2017 ◽  
Vol 3 (2) ◽  
Author(s):  
M. Benz ◽  
T. Schulenberg

A new numerical model for stratified two-phase flows with wavy interface is derived in this study. Assuming an equilibrium condition between turbulent kinetic energy, potential energy, and surface energy, the turbulent length scale in the inner region of a two-layer turbulence approach can be described by a statistical model to account for the influence of the waves. The average wave number, which is an input parameter to this model, is taken from wave spectra. They can be predicted from a Boltzmann statistic of turbulent kinetic energy. The new turbulence model is compared with the two-phase k–ϵ turbulence model. Time-averaged flow properties calculated by the new approach, such as velocity, turbulence, and void profiles, are shown to be in good agreement with experimental data.


Sign in / Sign up

Export Citation Format

Share Document