|
|
|
| Untersuchte Arbeit: Seite: 209, Zeilen: 19-32 |
Quelle: Ko and Ryou 2005 Seite(n): 1304, Zeilen: 5-14, 16-17 |
|---|---|
| However, there are at least three limitations in the Fluent model. The first is that the Fluent model assumes that a given parcel may collide with another parcel only if these two parcels lie in the same computational cell. As indicated by Schmidt and Rutland (2000) and Nordin (2001), this assumption may be inappropriate. Under this assumption, the collision between two spatially very close parcels is a priori ignored if they reside in different computational cells. Contrary to this, the collision may occur for a pair of possibly far distant parcels in the same computational cell. As a result, the collision model strongly depends on the computational cell sizes. The second limitation is linked with non-uniformity of the spatial distribution of parcels in the domain. Aneja and Abraham (1998) indicated that the Fluent approach is not suitable for sprays, where the variation in number density is large even inside one cell. Finally, the Fluent model considers only two collision regimes such as separation and [permanent coalescence.]
Aneja, R. and J. Abraham (1998). How far does the liquid penetrate in a Diesel engine: computed results vs. measurements? Combust. Sci. and Tech. 138, pp. 233-255. Nordin, P.A.N. (2001). Complex Chemistry Modeling of Diesel Spray Combustion. Ph.D. Thesis, Chalmers University of Technology, Sweden. Schmidt, D.P. and C.J. Rutland (2000). A new droplet collision algorithm. Journal of Computational Physics 164, pp. 62-80. |
However, there are at least three limitations in the O’Rourke model. The first is that the O’Rourke model assumes that a given parcel may collide with another parcel only if these two parcels lie in the same computational cell. As indicated by Gavaises (1997) recently, this assumption may be inappropriate. Under this assumption, the collision between two spatially very close parcels is a priori ignored if they reside in different computational cell. Contrary to this, the collision may occur for a pair of possibly far distant parcels in the same computational cell. As a result, the collision model strongly depends on the computational cell sizes. The second limitation is linked with the ignorance of the preferred directional effects of droplets. Nordin (2000) has indicated that the O’Rourke’s approach is not suitable for sprays, where the variation in void fraction is large even inside one cell and the parcels have the so-called ‘preferred directions’. [...] Finally, the O’Rourke model considers only three collision regimes such as separation, permanent coalescence, and grazing bounce.
Gavaises, M. (1997). Modeling of diesel fuel injection processes. Ph.D. thesis, Imperial College of Science and Technology and Medicine, Department of Mechanical Engineering, University of London. Nordin, N. (2000). Complex chemistry modeling of diesel spray combustion. MS. thesis, Chalmers University of Technology, Sweden. |
Not marked as a citation. Interestingly enough, although the model used is different, the text is mostly taken verbatim. At the end the text modifications needed to become more massive. |
|