'Wing structure FEA analysis' simulation project by hsturrock1


I created a new simulation project called 'Wing structure FEA analysis':

FEA analysis on internal wing structure.

More of my public projects can be found here.


Hi @hsturrock1,

Cool wing design with very detailed structural elements !

I see you are stuck with the mesh operation

I would like to suggest few things to consider to mesh such a detailed/complicated geometry.

Few years ago, I tried automeshing meshing the entire wing-tail assembly in ANSYS and ran into this

I realized geometry had to be cleaned up and some structural element had to be meshed separately. Auto-mesh algorithm does wonders no doubt, but when it comes to detailed structures it crashes or might take several hours to compute.

My suggestions would be:

  • Try to eliminate the wing stringers (small rods), the reason is that this structure would not take the maximum flight loads, and the load path does not get altered by removing it (Although exact stress results may not add up)

  • The skin must be removed since it is a very thin structure (From your model it appears to be nearly zero thickness). Thin structures/plates have to be meshed using shell elements or possibly higher order elements. If you are using a stressed skin concept for your wing design then the skin is supposed to be thick

  • Use local mesh refinements for critical regions like I-beam and holes

  • Sometimes the thin structures like the edge of trailing section might cause trouble, a good practice is usually to give some tolerance for trailing edge during CAD modelling.

  • Meshing the internal regions of the wing is very tricky and time consuming. I would suggest you to just re-modify the things in I-beam such as the fillets as these are very tiny and difficult to catch an eye

I guess with these suggestions, you could reduce the complexity of geometry and make it as simple as a rib-spar assembly as shown below

With this meshing would be much easier, although it might take a lot of time as the model is quite huge.

In any case it will be interesting to see the cleaned model ready for meshing.

hope this helped.
Cheers !