Hello, I am a mechanical engineer student. The subject of my research is to study the mechanical response of tensile probes with different filling patterns (Hexagonal honeycomb, Triangular honeycomb, Square honeycomb) and then compare the results of the simulations to define the better option for an optimal strength/weight ratio for a 3D printed part. 3D printers widely use the different infill patterns to save material and weight, but little is known about which type of infill is most appropriate for each type of load. My university's laboratory wanted to carry out experimental tests to validate my simulation results, but unfortunately, due to the Covid crisis, these tests will not be carried out. I wanted to ask you if there would be another kind of validation I could do without doing experimental tests. Or if there is any way to prove that my simulation results are valid.
All specimens will be tested to a Tensile stress test in Ansys WB, where I defined the same material and boundary conditions. The material is PLA. I leave images to illustrate the boundary conditions used in the simulation. The specimen's right end has a displacement of 2 [mm / min], as seen.
I would also like to ask you if you think these are the correct boundary conditions. Because I had incorrect results with other conditions, and these seem to give acceptable results. For some reason I cannot explain when I was doing a mesh convergence study (reducing the size of my mesh elements), the values of the Stress never converged. For these conditions I used on my model, I did not experience this problem. If someone could explain to me why I had this problem, I would be very grateful.
Thank you for your reply kind Sir.
For the case of Verification, using another solver like COMSOL (a different code solver than Ansys) and solving the same system, with the same boundary conditions (along with the mesh convergence study) and getting the same results would be a way to verify my results?
For the case of Validation, since I cannot do experimental tests and I couldn't found experimental data on my simulation case from other authors, the only way of validating my results would be using hand calculations?
U need to develop an mathematical model describing the problem and validate it.
I like the distinction between verification and validation cited by Mr. Anas . In this wording a hand calculation is a verification because if it solves the same equation as is done by an alternative FEM solver. In this definition a validation can only be achieved by experiments. But the definition assumes perfect conditions.
In less perfect situations, the verification by hand calculation can only be done approximately because hand calculation is done with the solution of a simplified problem. Also, experiments are less controlled than numerical analysis and contain uncertainties. In a weakened sense you could do a validation by analysis if you are not sure about the effect of your model assumptions: add the neglected effects to your model and check their influences. This however, is not a strict validation because you never know if you have got all effects with their proper data.
- Badges
- Science topic
- Similar topics
- Computer Communications (Networks)
- Simulators