Impact of two deformable meshes with velocity

Hi everyone. I’m working on a simulation problem for ballistic impact on a material that I want to simulate in the viso-elastic regime. I’m new to FEniCS and FEM in general, coming from a background in semiconductors simulation. I found a number of helpful examples for stress/strain simulation (Solving PDEs in Python - <br> The FEniCS Tutorial Volume I), and one for simplistic contact (Hertzian contact with a rigid indenter using a penalty approach — Numerical tours of continuum mechanics using FEniCS master documentation) but not quite something similar to what I need. I ran through those to catch myself up as best I can.

What I’d like to do is setup two deformable meshes (the projectile and the target) with their own material parameters, use a governing equation to determine how the impact deforms the two shapes (was thinking of Johnson-Cook), and then find the result of a given impact velocity with given materials. Do you recommend any resources or examples of two deformable meshes coming into contact in FEniCS? Something where I can include friction and poission ratio due to compression? Am I using the right tool for this problem?

Thank you.

Hi, contact simulation between two deformable meshes is really something that FEniCS has not been designed for. That’s why contact simulation that you can find using legacy FEniCS are overly simplistic.
Although I didn’t give it a try yet, you may be able to achieve something using the multiphenics project or the MultiMesh functionality. It would however certainly need some work.

1 Like

FEniCS is not specifically designed for anything other than offering the tools to assemble bespoke finite element matrices. This offers incredible power and flexibility to implement any FEM based numerical scheme you can imagine; however, at the cost of complexity and implementation work. This leads to great popularity in research communities who need fine control over all components of their algorithms.

The contact problem you suggest is very difficult. It is entirely possible to implement this using FEniCS, but requires expert knowledge of many areas of numerical analysis, computational physics and computer science. For example @dokken is currently doing work in this area.

Typically if someone is looking for a “black box” for a physics simulation, I’d recommend commerical packages.

1 Like

Thank you. That is what I anticipated. And while I revel in the challenge to do that, my expertise isn’t in mechanical modeling so I’m afraid it would take me quite a while to catch up, and time I don’t have.

1 Like