HobbyKing.com New Products Flash Sale
Reply
Thread Tools
Old Dec 08, 2014, 11:20 AM
Sink stinks
Montag DP's Avatar
United States, GA, Atlanta
Joined Apr 2005
4,597 Posts
Quote:
Originally Posted by martig View Post
As far as I've understood camber and thickness are reduced to have better performance at low Reynolds. I think flow separation is the issue.

I used 4 operating points - 0, 0.3, 0.6, 0.8 and just scaled the root's Reynolds to match.
What happens when you run the final sections at different Reynolds numbers? For example, take the root section and run it at the Cls you optimized for, but with the tip Reynolds number. Does it perform better than the optimized section for the tip? If so, then it means the optimizer is having trouble optimizing the tip section. If not, then it's doing its job. You can also try the reverse, testing the tip airfoil at the root Reynolds number.

It does seem strange at first that the camber would increase going out towards the tip, but maybe it actually makes sense. If the lower Reynolds number causes a decrease in Cl, all else being equal, then to get the same Cl, either the angle of attack needs to be increased or the camber. It seems reasonable that the shape could be tailored by the optimizer to increase camber while still minimizing flow separation.
Montag DP is offline Find More Posts by Montag DP
Last edited by Montag DP; Dec 08, 2014 at 11:27 AM.
Reply With Quote
Sign up now
to remove ads between posts
Old Jan 02, 2015, 10:57 PM
Registered User
spoke2570's Avatar
Florida
Joined Aug 2005
18 Posts
I have questions about basic use of this software.

I have been using Xfoil for a bit now. My mac is operating on Yosemite. X11 is installed. XoptFoil 1.3.1, the included, minimal version of Xfoil, and Design Plotter seem to be working properly.

After XoptFoil, and Design Plotter run, I am given 5 .dat files: design_coordinates, design_polars, optfoil, particleswarm_designs, simplex_designs.

What software should I be using to graphically view these data?
What must I do to extract a particular iteration of foil redesign for evaluation by Xfoil, or just to plot it?

Feeling kinda dense here... HELP
spoke2570 is offline Find More Posts by spoke2570
Reply With Quote
Old Jan 03, 2015, 12:39 PM
Sink stinks
Montag DP's Avatar
United States, GA, Atlanta
Joined Apr 2005
4,597 Posts
Quote:
Originally Posted by spoke2570 View Post
I have questions about basic use of this software.

I have been using Xfoil for a bit now. My mac is operating on Yosemite. X11 is installed. XoptFoil 1.3.1, the included, minimal version of Xfoil, and Design Plotter seem to be working properly.

After XoptFoil, and Design Plotter run, I am given 5 .dat files: design_coordinates, design_polars, optfoil, particleswarm_designs, simplex_designs.

What software should I be using to graphically view these data?
What must I do to extract a particular iteration of foil redesign for evaluation by Xfoil, or just to plot it?

Feeling kinda dense here... HELP
Hey Spoke,

No, you're not being dense at all. The files are written in a format that can be loaded directly and displayed in a program called Tecplot 360, which we have where I work. I realize most people don't have that, but you can still use the files. Here is a description of each of them.

particleswarm_designs.dat: Values of the design variables for each subsequently better design found by particle swarm optimization.
simplex_designs.dat: Same as above, but for the direct search algorithm.

You will get the above two files if your search type is "global_and_local." Otherwise, you will just get one of them. Those files act as input files for the design_plotter tool; they are not actually meant to be viewed graphically by themselves. The design_plotter tool takes them and uses them to generate airfoil coordinates and polars.

design_coordinates.dat: This file is generated by design_plotter. These are the x, y coordinates of each subsequently better airfoil design found by the optimizer. They don't exactly correspond to optimizer iterations, because during some iterations the airfoil design isn't improved. But if you want to analyze one of these not-fully-optimized airfoil designs, just copy and paste the coordinates under one of the "zone t=..." labels. You can graph these in Excel, Matlab, etc., or use them to make a .dat input file for XFoil. You could also write a script in Python or Matlab to plot them automatically, if you know how to do that.
design_polars.dat: The lift and drag coefficients for each subsequently better design found by the optimizer, corresponding to the coordinates in design_coordinates.dat, at the operating points you requested for the optimization.
optfoil.dat: The coordinates of the final optimized design. These are the same as the coordinates of the last design in design_coordinates.dat, but this file is written automatically by the optimizer at the end of a run (you don't need to run design_plotter to get it).

Hope this helps!
Montag DP is offline Find More Posts by Montag DP
Reply With Quote
Old Jan 03, 2015, 09:17 PM
Registered User
spoke2570's Avatar
Florida
Joined Aug 2005
18 Posts
Thanks! That helps a ton.

I am going to add that post to the instructional document that was supplied with your software. ..and I am going to suggest that you add the content of this post to that document, as well.

back to geek' out.
spoke2570 is offline Find More Posts by spoke2570
Reply With Quote
Old Jan 04, 2015, 03:52 PM
Sink stinks
Montag DP's Avatar
United States, GA, Atlanta
Joined Apr 2005
4,597 Posts
Quote:
Originally Posted by spoke2570 View Post
Thanks! That helps a ton.

I am going to add that post to the instructional document that was supplied with your software. ..and I am going to suggest that you add the content of this post to that document, as well.

back to geek' out.
Definitely... I don't think I spelled out what those files do clearly in the documentation. Thanks for the suggestion.
Montag DP is offline Find More Posts by Montag DP
Reply With Quote
Old Feb 15, 2015, 02:22 AM
Registered User
Paul_BB's Avatar
France, IdF, Lagny-sur-Marne
Joined Feb 2007
622 Posts
Thanks you for sharing your amazing work.
Paul_BB is offline Find More Posts by Paul_BB
Reply With Quote
Old Feb 15, 2015, 04:11 AM
Registered User
Paul_BB's Avatar
France, IdF, Lagny-sur-Marne
Joined Feb 2007
622 Posts
Did you ever think of using spline points as airfoil variables for your optimization?

When I play with XLFR5 (XFOIL) I am amazed at how accuratly a spline defined by 6 points (6 for the upper airfoil and 6 for the lower airfoil) can define most airfoils.

Would it be interesting to do so in your program? The coordinates of the spilne would be the vector of 2 x 6 x 2 = 24 coordinates?

Just my two cents...
Paul_BB is offline Find More Posts by Paul_BB
Reply With Quote
Old Feb 15, 2015, 02:56 PM
Sink stinks
Montag DP's Avatar
United States, GA, Atlanta
Joined Apr 2005
4,597 Posts
Hey Paul,

I did try using spline control points as design variables before in my original airfoil optimization tests with Matlab. It may have just been the type of spline that I was using, but it didn't seem to work as well as the perturbation methods I am using now with Xoptfoil. Thanks for the kind words about the program.

Dan
Montag DP is offline Find More Posts by Montag DP
Reply With Quote
Old Feb 16, 2015, 12:14 AM
Registered User
Paul_BB's Avatar
France, IdF, Lagny-sur-Marne
Joined Feb 2007
622 Posts
Dan,

I have spent half of my professional life optimizing engineering solutions with the classic methods and I know things never go as you want. And I can also imagine the amount of work you have already put in there, not to mention the pain and frustrations. Keep up the good work!
Paul_BB is offline Find More Posts by Paul_BB
Reply With Quote
Reply


Thread Tools

Similar Threads
Category Thread Thread Starter Forum Replies Last Post
Discussion Airfoil Optimization Using Genetic Algorithms Ghost_ Modeling Science 38 Apr 24, 2014 07:18 AM
Discussion Airfoil Optimization with XFOIL Montag DP Modeling Science 168 Feb 03, 2014 09:14 PM
Discussion Evolutionary airfoil design optimizer available kcaldwel Modeling Science 3 Dec 10, 2012 05:03 PM
Profili - Xfoil and reflexed / autostabilizing airfoils surfimp Modeling Science 29 Mar 30, 2005 07:32 AM