Amplification factors in SolidWorks Flow Simulation?
 

I am trying to use Flow Simulation for analysis of Amplification factors, due to wind impact on my building surroundings.


The "amplification factor" is defined as the ratio of the local wind speed to the wind speed that would occur at the same position without the buildings present.

By this "without the building present" I do not mean that before the building was built, there was a flat plain on that location, without anything on it. It simply means, that a new object (building) has been built somewhere on the location, where the other objects were already present.


Here is an example of the amplification factor analysis in some other CFD application (I do not know which one):

http://imageshack.us/photo/my-images/59/53940521.jpg/

So it seems this is possible to be done is some CFD application.


Can anybody tell me if this can be done in SolidWorks Flow Simulation?
Or maybe somehow by getting the separate results from SolidWorks Flow Simulation:

http://imageshack.us/photo/my-images/856/25347812.jpg/

and then editing it in Photoshop to get the Amplification factor?


Any help please?
Thank you

Hi George,
careful with photoshop and scientific research. I don't know how the amplification factor is calculated but I assume you compare the two results in an equation that gives you the distribution for a third image. Similar like:
(Result B - Result A)/Result A=Result C
where A is without building, B with building and C the amplification factor.
This cannot be done in FloEFD as you cannot calculate with results between two other projects. What you could do is use a point parameter grid distribution for the values you need of both projects and work with them in Excel where you can calculate the amplification factor. But it's not pretty.
If you have Techplot you can readin excel files and visulize them. I did this once with electronically recorded measurments which I also had to apply in a calculation in Excel and then visualized it in Techplot. But I don't suppose you have Techplot :-)

Hope this helps,
Boris

Thank you for the reply Boris.

I can download a Trial version of Tecplot 360:
http://www.tecplot.com/Downloads/Eva...ecplot360.aspx

Can you be a more specific? I need to do a point parameter grid distribution? How to do this in SolidWorks Flow Simulation?

"Point Parameter" is a post processing feature where you can either select predefined points from the geometry, define coordinates or select a reference geometry or plane and define either by spacing or by number. Select the parameter you want and then export it to Excel.
For Techplot, I don't know the newest version and it's too long ago to remember how I did it, but as far as I can remember you have to use the resulting numbers only, no cells with equations in it and it has to have a plain format if I remember right. So first row were one axis values (like x-axis: 0; 0.1; 0.2 etc.) and the first column the other axis values, without any labeling. But I'm not sure anymore. The rest I think you are able to select in Techplot during import or so.

Boris I do not intend to contradict your explanation, just I am confused with this:

I do not understand why values for a particular axis (x in this case) are in one row, and then the other values for the other particular axis are in the other column?

I guess it would be logical that axis values are in each row, or column.

Or just I have not understand you?

In your example of the other CFD image it is a 2D field of values. For example 100x100 values of the cells. So it's like a matrix of 100x100. 100 columns and 100 rows and you will need to tell at which coordinate they are. In your model the plane might be the x-y-plane. So you have the x coordinate of your point and the y coordinate of your point. Just like in a diagram and the value it selfe of your magnitude factor will be the amount of the corresponding x and y position which you can compare with a hight at certain coordinates on a map.
But as I said, I'm not quite sure anymore so you have to try yourself.

Now I understand. Thank you Boris.

Do you have a copy of that excel file of yours?

And what about the CFD analysis. In which application did you do it?

No, sorry. That was about 5 years ago for my thesis at that time.

Ok, thank you.

Can I contact you by chance, if I will have some difficulties regarding this issue?

If it's about FloEFD or SWFS you can write here and someone can help you if I don't do it first and about Techplot I would prefere you checking other forums as I haven't used it in a long time and don't have it. So you will most likely get more and better help from a Techplot forum.

Understood. I will post the problem in here.

Yes, its about the SolidWorks Flow Simulation (FloEFD is the plugin SolidWorks uses for CFD analysis?)

Yes, Mentor Graphics is the developer of FloEFD which is SWFS but in this case sold by SW. FloEFD is available as standalone, Catia V5, Pro/E and NX integration just like it is integrated in SW, 100% look and feel of the CAD system it is integrated and you'll never leave the CAD system just like in SW.

I understand.
This is probably the reason why nobody from the SolidWorks official forum, replied to this question.

I was better to post it at the FloEFD forum, if they even have one.
But you helped me a lot. I already made calulation regardin 10x10 grid point cell, now I am trying to find out if I can import these results into Solidworks again, and get the plot analysis. But it seems I am not succeeding.

Hi George,
There is a community for FloEFD but you will have to be customer at Mentor Graphics to get there. Other than that there is this forum here.

No, you won't be able to get results back into SWFS or FloEFD. You can only load result files but since ther is no possibility to compute a thrid result out of two other results and then load that back into SWFS/FloEFD you wont be able to do something like that which is why I would use Techplot. Yiou can do such kind of 3D diagrams also in other tools similar to excel but they usually also cost money or if there is a trial version they are limited in the capabilities such as amount of data etc.

I guess owning the Flow Simulation from SolidWorks, does not necessary counts me as a Mentor Graphics customer?



I am manually going to edit both two excel files, so that I get the third one (with "Result C"), and then load it back into SW. Is this possible?

How to do this, if it is? Load the excel file back to SW?

No, SW customer does not make you Mentor Graphics customer for support.

No, you cannot get any results back into SWFS to visulize.

Ok, I will try to do something, in Techplot.
If I experience any issues, I post it here.

Thank you once again. I really appreciate all you have done for me.

Hm, it seems that Point Goals can not be set to the points in the air, only on the points of solids (solid edges).

I tried to set some Goal points in the Flow Simulation, so that I can get Velocities for each of these points.

But for some reason, I can not do this.

When I try to set them in the "Reference points" mode, I am getting the following message:

"Cannot find vertex associated with this point goal."

http://imageshack.us/photo/my-images/833/23137941.jpg/

http://imageshack.us/photo/my-images/51/39329314.jpg/


Not to mention that it took my 4 hours of just clicking (around 5000 points) and selecting the points.



Then I tried to set the points, in the "Point Coordinates" mode, but again, I could not do this, because I am getting an error message:

"Please specify only solid parameters for this point"



What seems to be wrong? I can set up the points on solids (solid edges) but I can not use the points in the air? Why not?

by the way, this is what I mean by "Reference points" and "Point Coordinates" modes:

http://imageshack.us/photo/my-images/196/66361723.jpg/

Hehe, sorry for the 4hrs but "Point Parameter" is a postprocessing feature just like surface and volume parameter and you can define a offset plane (reference plane) which you select for the point parameters and select either by spacing or number of poins and then they will be distributed on this plane in the whole computational domain.

Here's how it looks like:
http://imageshack.us/photo/my-images...parameter.png/

Thank you for the help Boris.

It seems I am not doing something right.

When I apply the Point parameters to my "Plane 3":

http://imageshack.us/photo/my-images/706/50385678.jpg/

They are distributed all around the project, on the same height like the "Plane 3":

http://imageshack.us/photo/my-images/13/55799037.jpg/

I thought that by shrinking the "Plane 3" dimensions (in x,y directions) I would allow the Point parameters to be distributed only inside the "Plane 3". But it seems not.

How can I distribute the points only into particular part of the project?

A plane is not defined in size, just in location, therefore it is distributed on that plan in the whole computational domain. If you want to specify a certain area you will have to create a part/body with a planar surface on that level you want to have the points and select this surface instead of the plane. then the points will be distributed only on the surface as a surface is limited by its size.
But make sure you deactivate the part/body in the component controls as it is not a solid in the calculation, it shall be not present for the calculation, just as a reference object.

I succeeded in getting the point parameters, results in excel file.
I will let you know if I have any problems with Techplot when I download it.

Thank you once again Boris. What would I do without you.

I know you already told me that you did it 5 years ago, but maybe you can recall something:

Here are the points I took for calculation from SolidWorks:

http://img853.imageshack.us/img853/5417/74396465.th.jpg

Here is the excel file I want to import into Tecplot 360, and get the plot from it:

http://img51.imageshack.us/img51/2197/47432988.th.jpg

And here is the result which I get when I import it:

http://img850.imageshack.us/img850/5365/55736240.th.jpg

Now, it seems I need to do some changes in the "Mapping Style"?

Here are all Mapping Style options:

http://img5.imageshack.us/img5/6931/52923541.th.jpg

Did you remember anything?

Hi George,
I'm sorry but I cannot really remember what the mapping style will change in the plot. But for me it looks like you used a x-y line plot as you can see in your third image.
You should try to rearrange the excel field with x values as row and y values as column.
I just searched techplot in google and got onto this page via Wikipedia:
http://download.tecplot.com/tecio/

Here you'll find a user manual which will help you much better than I can. Check out page 78 which is exactly what you'll need. Having 2D plot of your date from a certain Excel file format. as I said x coordinates are the first row, y coordinates the first column and the corrensponding values to these coordinates are in the intersecting cells to these coordinates.

Boris

Thank you for the help Boris. I succeeded! But the results do not look like they should.

I imported the .xls file into Tecplot, Created the Rectangular Zone, Interpolated the data from .xls file (this is all from the Help file, I think I did it correct).

Here is what I got as a result:

http://img843.imageshack.us/img843/4194/93328721.th.jpg

Here are the results I got in SolidWorks Flow Simulation, for wind velocities.
Without the building (curved roof):
http://img849.imageshack.us/img849/400/62289833.th.jpg

And with building:
http://img855.imageshack.us/img855/6644/52866567.th.jpg

So it seems something is wrong with the Aplification factors in Tecplot???

I don't know, have you checked the numerical data at the positions you think are wrong? If the red region is wrong but in the excel sheet the show the highest values than Techplot is correct and the calculation in Excel is wrong.

I did not understand you this part:

"If the red region is wrong but in the excel sheet the show the highest values than Techplot is correct and the calculation in Excel is wrong."

Up, sorry my quick typing.
I meant if the red area in your Techplot image is not correct but the Excel sheet values have the highest values at these coordinates then the Techplot results are good and you should check the excel calculations.
Simply compare the numerical values with the graphical ones. if the error happens in Techplot then there is a problem with the data translation but if Excel has already the highest values then you probably have the error in the excel calculation.

What seems to be strange is that I checked it 2 times, and both the data from Excel and Tecplot were correct.

I tried to lower the number of Points in the Point Parameters from 1000 to 100.
And it seems I succeeded to get the correct result, or at least it looks to me that this result is more correct than the previous one:

http://img825.imageshack.us/img825/899/63742200.th.jpg

Some points (especially the ones that were on the solid (roof)) had very low Velocity value presented in exponential way (example 2.3412121E-17).
What I did is converted all this values to 0.1 and then divide it between each other (values of velocity with and without the buildings).

Its interesting that I did the same thing at the 1000 points version.

So I do not know what seems to be the problem at the previous 1000 version?

Anyhow, I think this is the final solution? At least it looks like.

Maybe it's related to the mesh resolution. I cannot really tell.
You can also try to use a sketch on the surface you use for the points and draw a line through a fewsections and use them in a xy-plot to see the changes along this line to check if they are correct. Of course use the calculation of the parameters along the line again to calculate the amplification factor.

Thank you for the reply.

Again Boris I did not understand you. I am apologizing :(
Is this what you meant:

http://img828.imageshack.us/img828/3950/98675362.th.jpg

Uploaded with ImageShack.us

Sorry for ignorance :(

Cannot tell what type of line this is from the image.
Basically draw a line in a sketch on the surface you want to have the results. Then use the SWFS feature "XY-Plot" select the line and the parameter you need and export that into Excel, do the same for the other results and then calculate the magnification factor and plot that in a excel graph just like the single XY-Plots were plotted.

Here are the results I got from the XY Plot in Solidworks:

http://img199.imageshack.us/img199/8155/33403438.th.jpg

And here are the ones I got from making a section on the Tecplot contour:

http://img508.imageshack.us/img508/6293/55900790.th.jpg


So it`s OK?! :D

Thank you Boris. I owe you a favor! A beer, or couple of those!
You are truly invaluable member of this community.

THANK YOU!