Comparison between Direct Numerical Simulation and k–ε–Model Prediction of the Flow in a Vessel Stirred by a Rushton Turbine
Publication date
2000
Document type
Book part
Author
Editor
Akker, Hendrikus E. van den
Derksen, J. J.
Organisational unit
Conference
10th European Conf. on Mixing, Delft University of Technology, The Netherlands, July 2–5, 2000
Book title
10th European Conference on Mixing
First page
239
Last page
246
Peer-reviewed
✅
Part of the university bibliography
Nein
Abstract
The k-ɛ model is the most prevalent turbulence model in the numerical simulation of stirred vessel flows. Despite the fact that many other turbulence models—such as the κ-ω model and the RNG κ-ɛ model—have been introduced to improve the κ-ɛmodel, it has remained the most frequently utilized model for routine investigations in industry. The reason for this predominance is that the κ-ɛ model has been quite successfully applied to many complex flow problems. Moreover, the weaknesses of the model are relatively well known and can be accounted for. The comparison of the large-scale flow field shows a good overall agreement of the observed flow features for both types of simulations. On the other hand, close comparison revealed differences between DNS and κ-ɛ simulations where the DNS predictions were almost always closer to experimental results. The results for the turbulent kinetic energy κ also show some evidence of better agreement between experiment and DNS but it is not completely clear right now how reliable the information for turbulent kinetic energy is.
Cite as
Bartels, C., Breuer, M., Durst, F.: Comparison between Direct Numerical Simulation and k–ε–Model Prediction of the Flow in a Vessel Stirred by a Rushton Turbine, Proc. of the 10th European Conf. on Mixing, Delft University of Technology, The Netherlands, July 2–5, 2000, H.E.A. van der Akker and J.J. Derksen (eds.), pp. 239–246, Elsevier Science B.V., Amsterdam, (2000).
Version
Not applicable (or unknown)
Access right on openHSU
Metadata only access