CFD Online Logo CFD Online URL
www.cfd-online.com
[Sponsors]
Home > Forums > General Forums > Main CFD Forum

TVD

Register Blogs Community New Posts Updated Threads Search

Reply
 
LinkBack Thread Tools Search this Thread Display Modes
Old   March 7, 2002, 15:47
Default TVD
  #1
nu
Guest
 
Posts: n/a
Hi

what do we mean by a TVD schemes. is something to deal with limiter functions or the numerical solution is nonincreasing in time ?

any numerical help!
  Reply With Quote

Old   March 7, 2002, 18:02
Default Re: TVD
  #2
ryoga
Guest
 
Posts: n/a
TVD stands for total variance diminishing. A good introduction into TVD schemes and flux limiters, as well, is provided in C. Laney's book, Computational Gasdynamics. Hope this helps!

Ryoga
  Reply With Quote

Old   March 7, 2002, 21:33
Default Re: TVD
  #3
Paul
Guest
 
Posts: n/a
As Ryoga said, TVD means Total Variance Diminishing. In computational, it usually means that your numerical scheme won't produces more extremes than last time in the evolution, thus your result will be free of spurious oscillations. A effective means to ensure TVD is by limiter, which can be exerted in the function or its derivative. A minmod function is usually involved to kick out the too much variation.
  Reply With Quote

Old   March 8, 2002, 04:52
Default Re: TVD
  #4
nu
Guest
 
Posts: n/a
i know what stand for, i've started my numerical form Toro's book. thanks paul. i.e. what i wanted to know, well done.
  Reply With Quote

Old   March 11, 2002, 11:50
Default Re: TVD
  #5
Bert Laney
Guest
 
Posts: n/a
Perhaps the best way to define the "TVD" property is as follows: the sum of the maxima in the solution minus the sum of the minima does not increase in time. Strictly speaking, "TVD" does not say anything about invidual maxima or minima -- it only affects the sum total of extrema.

"TVD" is best understood as a type of stability condition. The most common symptom of numerical instability is spurious oscillation, especially of the sort that grows progressively larger with time, kind of like a car with worn shocks on a bump road. "TVD" limits the size of but does not prevent spurious oscillations. Most so-called "TVD" enforcement techniques actually ensure something much strong than "TVD," and tend to strongly discourage spurious oscillations (except possibly those that arise through interactions between characteristic variables, such as at shocks and reflective boundaries).

"TVD" is often enforced using techniques such as flux limiters, artificial viscosity, self-adjusting hybrids, etc.

A signature side-effect of "TVD" is to "clip" the peaks of maxima and minima, reducing the order of accuracy there or even inducing an artificial flat spot.

"TVD" is most often used for high-speed flows and especially shock capturing methods. A proper shock capturing method is conservative. Numerical conservation plus "TVD" implies more than either alone -- the combination strictly limits the order-of-accuracy at extrema to one.

There is obviously a great deal more to be said but hopefully this is enough for an internet chat group.

Bert Laney
  Reply With Quote

Reply


Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are On
Refbacks are On


Similar Threads
Thread Thread Starter Forum Replies Last Post
Nvd / tvd MichiB OpenFOAM Programming & Development 8 July 21, 2021 05:27
TVD ... Questioner Main CFD Forum 1 June 7, 2006 08:54
4th and 5th Order TVD Runge-Kutta Methods saygin Main CFD Forum 2 January 30, 2006 11:45
TVD TVD Main CFD Forum 0 November 17, 2005 11:25
TVD Schemes Matt Umbel Main CFD Forum 1 January 25, 2000 04:21


All times are GMT -4. The time now is 23:39.