CFD Online Discussion Forums

CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   Main CFD Forum (https://www.cfd-online.com/Forums/main/)
-   -   multiplicity of solutions (https://www.cfd-online.com/Forums/main/14618-multiplicity-solutions.html)

student January 11, 2008 04:24

multiplicity of solutions
 
Dear Sir,

I am new to Cfd . I have seen a cavity problems produces multiple of solutions. What does it mean ? can any one explain this ? many thanks .

Giovanni January 11, 2008 07:30

Re: multiplicity of solutions
 
Hi, as You know N-S equations are non-linear PDE, insofar these lead to significant mathematical problems about existence and uniqueness of the solution, in particulary up today no one was able to proof any kind of general theorem about uniqueness of N-S equations'solution. So different discretization scheme ad turbolence modelling choice can lead to different solution, CFD analisys is affected by this, and You can trust your CFD results only after an accurate validation process, involving check of convergence, the physical sense, (i.e. if your fluid flows from low pressure to high pressure zone you have some problem...) and possibly the agreement with available experimental data and empirical correlations.


A.K. January 11, 2008 08:48

Re: multiplicity of solutions
 
Sometimes, the multiplicity of numerical solutions is caused by a bifurcation of exact solutions to the N-S equations, irrespectively of the discretization. In such cases, multiple numerical solutions simulate real flows which occur in actual practice.


All times are GMT -4. The time now is 16:24.