CFD Online Discussion Forums

CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   Main CFD Forum (https://www.cfd-online.com/Forums/main/)
-   -   conservative and non conservative (https://www.cfd-online.com/Forums/main/12011-conservative-non-conservative.html)

jj August 15, 2006 11:25

conservative and non conservative
 
dear sir, can we say eulerian approach is conservative form and lagrnagian approach as non conservative form?

ag August 15, 2006 14:07

Re: conservative and non conservative
 
No - an Eulerian approach can be non-conservative or conservative. A conservative approach involves casting the flux terms as a divergence of a flux function.

Deepak August 16, 2006 00:50

Re: conservative and non conservative
 
ya any approach can be converted into conservative or non conservative forms..we can do this using mathematical formulations like using advection terms.but can anybody explain the physics going behind that? like how by just changing into divergence form,u can assure conservative?

Thanks in advance,

Deepak

ag August 16, 2006 07:51

Re: conservative and non conservative
 
The differential equations that we normally solve are actually derived from integral equations that hold globally. When they are reduced to differential form then the equations hold locally and we need extra conditions to ensure global conservation. The divergence form is one form that can be directly related to the integral form of the equations via Gauss' Divergence theorem. It thus possesses better conservation properties than a non-conservative scheme (for example the ability to capture shocks without the explicit application of the Rankine-Hugoniot condition). At least, that's how I learned it. More information could be found in the Wiki article, as well as googling for conservation form+CFD. Most textbooks also include some information on this.


All times are GMT -4. The time now is 08:00.