2015
DOI: 10.1016/j.jcp.2015.07.017
|View full text |Cite
|
Sign up to set email alerts
|

Truncation effect on Taylor–Aris dispersion in lattice Boltzmann schemes: Accuracy towards stability

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

1
14
0

Year Published

2018
2018
2021
2021

Publication Types

Select...
6

Relationship

1
5

Authors

Journals

citations
Cited by 27 publications
(15 citation statements)
references
References 55 publications
1
14
0
Order By: Relevance
“…As by default, the LBM evolves as a time-marching procedure, the algorithm can be initialized with A = 1 and so maintained until the simulation is about to reach the steadystate criterion; at this point, we switch to A = 0 so that, right after, when the steady-state regime is reached the corresponding time-independent solution supports the consistency properties discussed in this work. This procedure mimics the protocol employed in the flagging of the anti-numerical-diffusion correction popularized in the simulation of time-varying advection-diffusion problems [43,45] .…”
Section: Discussionmentioning
confidence: 99%
See 1 more Smart Citation
“…As by default, the LBM evolves as a time-marching procedure, the algorithm can be initialized with A = 1 and so maintained until the simulation is about to reach the steadystate criterion; at this point, we switch to A = 0 so that, right after, when the steady-state regime is reached the corresponding time-independent solution supports the consistency properties discussed in this work. This procedure mimics the protocol employed in the flagging of the anti-numerical-diffusion correction popularized in the simulation of time-varying advection-diffusion problems [43,45] .…”
Section: Discussionmentioning
confidence: 99%
“…This distinction is well-established within CFD community. For example, the numerical diffusion phenomenon, which originates from a mixed space/time derivative term of the advective flux, is known to affect time-dependent problems, where it can be cancelled with the inclusion of a anti-numerical-diffusion correction [42][43][44][45] . At steady-state this error does not exist, so the inclusion of such a correction, rather than beneficial, will introduce the exact same numerical diffusion artefact that was designed to eliminate in the first place.…”
Section: Introductionmentioning
confidence: 99%
“…28 The numerical simulations in this work will be performed with the steadystate formulation, which enables us to verify all new bounding techniques through a quite arbitrarily, parameter range (physical contrasts, P eclet number), without concerns from stability. 31,32,34,38,60,83 However, the transient MR ADE interface-conjugate has been also developped; 40,42 its update to N-MR is identical with the boundary counterpart, and it is exemplified with the steady-state interfaceconjugate treatment in Sec. III D. The MR interface-conjugate is also formulated 40 for the velocity and normal-stress continuous flow conditions; their release from the tangential stress projection shell becomes possible following the N-MR path.…”
Section: F Summary and Extensionsmentioning
confidence: 99%
“…1. the couple of the TRT bulk equations ( 26), assembled for every bulk link; 2. the local mass-conservation equation ( 28), assembled for every fluid node; 3. the MR closure equation (34), assembled for every wall-cut link or, alternatively, the A-LSOB closure equations assembled innode as provided below. 4. the couple of the MR interface-conjugate equations (36a) and (36b), assembled for every interface-cut link; they replace the couple of bulk equations.…”
Section: E Steady-state Algorithms With the Mr N-mr And A-lsobmentioning
confidence: 99%
See 1 more Smart Citation