Open Discussion: loads physical understanding in FEA

Yesterday, I got this message from an FEA engineer in Norway, Valeriy Shepenkov:

My biggest challenge and the biggest challenge for most in the finite elements business is the loads physical understanding and their application in the FEA software.
If you could provide some advice on how to effectively single out the essential loads and how to apply them to the structures in the FEA software, that would help at least to me to get better at FEA.
I hope my topic is of interest and that you will write a blog about it with your thoughts on the matter.

I thought this was an interesting topics and I asked more details to Valeriy and this is his answer:

I was interested in learning how to apply the loads, the advanced loads application. What is the general methodology on how to convert the physical loads into the FEA formulated loads with preserving the essential property of the load. Once talking to a professor from TU Delft, he also told me that this is one of the most challenging tasks in FEA to be able to decipher the loads meaning and be able to apply them in the FEM problem appropriately.

This example is from the subsea engineering:
For instance, you can think of a steel foundation structure (plate thickness 0.025 m, overall size 5×6 meters, skirt length 0.6 m) entering the sea splash-zone from a suspended in air position.There are resistance forces arising during the entering. So, how should we apply the loads in this case? Should it be a distributed pressure applied to the surface of the mat? What if we enter the splash-zone at angle? how do we need to apply the loads then? what if vertically?

Of course, this is an obvious challenge, to apply the loads. However, setting a clear mindset on how to do this properly will help to avoid mistakes in setting up the FEA model.

In my opinion, The challenge lies more in finding the correct way to model an engineering problem, than in the load itself.
A finite element model is in fact just a mathematical model and not necessarily a representation of the physic reality… It requires usually a lot of tests, iterations and different modeling approaches and of course some comparison with experiment or laboratory tests.

For example, you could model the structure and the water and simulate all that using Fluid structure interaction to see the effect of waves on the structure…but this would be very advanced analysis and in most of the cases you don’t need to go so far and you can just simulate as a pressure with probably a function which will model different pressure distributions.

I would be glad to have the opinion about other FEA Experts on this topic !

Write your comment in the comment section and I’ll add it inside the post