1
$\begingroup$

I can see lots of research articles around aerodynamic shape optimization of airfoils and wings like this MDOlab program. But all stuff like that don't seem to be available for a simple customer. Personally I've not seen where I can buy or download it. The only programs for that purpose that I've managed to find are:

  • Tosca Fluid (not adapted specifically for the airspace industry)
  • ANSYS Adjoint Solver (probably OK, don't know yet)
  • This Python code coupled with XFoil, but it's primitive and only for airfoils, not for wings.

Somebody know some accessible software for shape optimization of airfoils and wings for an ordinary engineer, but not for scientists or programmers?

$\endgroup$

3 Answers 3

5
$\begingroup$

If you follow the AIAA publication number (AIAA-2014-0567) given in the first frame of the linked video, you will find this article which explains in detail that the optimisation shown in the video was done on MACH:

These tools are components of the framework for multidisciplinary design optimization (MDO) of Aircraft Configurations with High fidelity (MACH). MACH can perform the simultaneous optimization of aerodynamic shape and structural sizing variables considering aeroelastic deflections.

If you now google for MACH, you will find this article by Gaetan Kenway which explains in detail how the software works. But this is not what you want, my impression is you need a simple to use software which can be downloaded and run within minutes.

My recommendation would be XFLR5 which incorporates the XFOIL airfoil solver and adds modules for designing and calculating the whole aircraft. YouTube has a variety of tutorials available. Is that what you want?

However, a basic understanding of aerodynamics and the possibilities and limitations of the algorithms used will be very helpful - a good engineer should be able to think both like a scientist and a programmer.

$\endgroup$
3
  • $\begingroup$ Do you know if MACH is a software for internal use only? I can't find any links to get it and no source codes. $\endgroup$
    – Bzhenko
    Dec 19, 2016 at 16:23
  • $\begingroup$ @Bzhenko: Yes, I could also not find any link - maybe you ask the MDO lab at the University of Michigan directly? But MACH is not something simple that can be used casually - it looks more like you need to work for weeks until you get the first case running. $\endgroup$ Dec 19, 2016 at 21:04
  • $\begingroup$ openmdao.org This website has an Open source tool for multidisciplinary optimization. $\endgroup$ May 26, 2018 at 8:09
3
$\begingroup$

I would recommend the use of Aeolus ASP for aerodynamic shape optimisation.

Edit: I am not associated with the software in any way. I had found its shape optimisation features useful for an assignment at my university.

$\endgroup$
3
  • 1
    $\begingroup$ In order to avoid this post to be marked as spam, could you please clarify: - are you associated with them? - why would you recommend that? $\endgroup$
    – Federico
    Jun 18, 2020 at 9:52
  • $\begingroup$ I am not associated with Aeolus, but their software was really useful and handy for shape optimisation for an assignment I had to do at my university $\endgroup$ Jun 18, 2020 at 9:54
  • 1
    $\begingroup$ Congrats! You're one of the very few who have ever responded to that prompt! Greatly appreciated. $\endgroup$
    – FreeMan
    Jun 18, 2020 at 16:47
2
$\begingroup$

I'd also recommend looking at XFLR5 because that's what I see most people in the homebuilt experimental and radio controlled communities use. Specifically, most competition planes like F5D pylon racers and F3K hand launched gliders are almost all designed on XFLR5.

But XFLR5 is primarily for simulation. Verifying your "hunches" are correct and testing your tweaks. Design and optimization is still mostly done by gut feeling and educated guesses.

However, the RC community have come up with an XFoil based genetic algorithm optimizer. Check out XoptFoil (google it). You just need to give it a starting airfoil and the target characteristics you're after and let it evolve it until it gets as close as possible to your target.

Note though that AI algorithms like the genetic algorithm can sometimes give nonsense results because it's the one that fits best to the desired target result. It only means that the algorithm have managed to generate the best airfoil that has the characteristics you want in the simulated world of XFoil, not necessarily the real world. You still need to build the wing and verify.

$\endgroup$

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .