Include Page |
---|
...
|
...
|
Include Page |
---|
...
|
...
|
...
Verification & Validation
Verification and validation is a formal process for checking results. Each of these terms has a specific meaning which we won't get into here. We'll act like consultants and do the bare minimum: we'll refine the mesh and make sure that the results are nearly independent of the mesh. We will also repeat the comparison with experiment which is a very important way to check simulation results.
Refine Mesh
Let's repeat the solution on a finer mesh with smaller cells. In workbench, under Forced Convection project, right click on Fluid Flow (FLUENT) and click duplicate. Rename the duplicate project to Force Convection Refined Mesh. You should have two project cells in workbench.
Double click on Mesh for Forced Convection Refined Mesh. The ANSYS Mesher window will open. Under Outline, expand mesh tree and click on Edge Sizing.
Highlight "Edge Sizing". Under Details of "Edge Sizing", increase Number of Divisions to 50. This will refine the mesh in the radial direction.
Highlight "Edge Sizing 2". Under Details of "Edge Sizing 2", decrease element size to be 0.02. This will refine the mesh in the axial direction.
Click Update to generate the new mesh. If you refer back to the mesh details and expand statistics, you will notice that the number of elements has increased to 15300, compared to 5508 of the original mesh.
...
Checklist
HTML |
---|
<iframe width="640" height="360" src="https://www.youtube.com/embed/odNQ8hHG7MA" frameborder="0" allowfullscreen></iframe> |
Check Boundary Conditions
HTML |
---|
<iframe width="640" height="360" src="https://www.youtube.com/embed/Fzrmdp_cS38" frameborder="0" allowfullscreen></iframe> |
Check Mass and Energy Conservation
HTML |
---|
<iframe width="640" height="360" src="https://www.youtube.com/embed/0ThDb-tVBQA" frameborder="0" allowfullscreen></iframe> |
Check Energy Conservation via Mixed Mean Temperature Variation (Optional)
HTML |
---|
<iframe width="640" height="360" src="//www.youtube.com/embed/YXkGqLC64-Q" frameborder="0" allowfullscreen></iframe> |
Check Numerical Error
HTML |
---|
<iframe width="640" height="360" src="https://www.youtube.com/embed/Xmc8fLUX3hc" frameborder="0" allowfullscreen></iframe> |
Validation Results
HTML |
---|
<iframe width="640" height="360" src="https://www.youtube.com/embed/fkZnLfcbLlc" frameborder="0" allowfullscreen></iframe> |
Finally, double click on Results cell of Forced Convection to compare. Under Outline tab, click on the results of interest to analyze (pretty sweet, huh?).
Centerline Temperature
...
https://confluence.cornell.edu/download/attachments/111221577/CenterlineTempComparison.png
Wall Temperature
...
https://confluence.cornell.edu/download/attachments/111221577/WallTempComparison.PNG
Axial Pressure
...
https://confluence.cornell.edu/download/attachments/111221577/AxialPressComparison.png
First Section Axial Velocity Profile
...
https://confluence.cornell.edu/download/attachments/111221577/FirstVelocityComparison.PNG
Second Section Axial Velocity Profile
...
https://confluence.cornell.edu/download/attachments/111221577/SecondVelocityComparison.PNG
Temperature Profile
...
https://confluence.cornell.edu/download/attachments/111221577/TemperatureProfileComparison.PNG
Comparison of Results
We can see from the charts comparing the results from the original and refined meshes that the results have changed very little and still compare well with actual simulation results. From this, we can say that our results have mesh-converged, refining the mesh further will not improve results.