It seems counter intuitive at first, but imagine an infinite version of this and look at some part. In which direction would it move (longitude wise)? None - because of symmetrical forces. Therefore it would seem that the only place you would have significant displacement would be at the free end.
So, nothing I have set wrong in the simulation...
Since thermal expansion=coefficient x beam length x temp.change , so I expect the max.displacement should be increased with the increase of beam length. But this simulation shows that no matter I increase the beam length, the max.displacement at the free end keeps not much different. Any maths principle can explain this result?
This is a reasonable result with the way you have your simulation setup.
The displacement is simply the amount of change from the original position. If you have a fixed geometry fixture holding the closest support, you're defining the displacement to be 0 at this location. The end of the beam is constrained by the fixtures and that is why the displacement is not proportional to the length of the beam.
The displacement will be proportional to the length of the beam if you eliminate the fixtures from the supports.
All the best,
what is the goal of your analysis?
what are you using as your test case? ie known results?
My purpose is to analyse the stress on every support due to thermal expansion(+30K).
I have created three models including
1000mm beam with 10 supports
2000mm beam with 20 supports
3000mm beam with 30 supports
I have probed the max. stress on every support and found an impressive result.
First, the end support has the max. stress (it is expected)
Second, no matter I increase the beam length, the max. stress is similar (mentioned above, the max. displacement at the end is similar too)
So, I believe this solidworks result but I am looking for some maths principle to explain it. Does Solidworks provides some maths principle about its FEA simulation to support its result?
validation examples are in the software
there is also the AFNOR document floating around
and also in 2014 there is a technical reference
all should give you confidence that the software is solving the problem correctly
but note, it is unlikely that you will find something that matches your case exactly
i would add that you should look at displacement before stress as stress is prone to singularities..etc. displacement is what is calculated by fea
Looks like you have everything set up correctly and I think the results are also correct.
Trying to figure out what's going on is an issue with everything you analyze. Running the
analysis is only the beginning, understanding the results is more work. Welcome to the
world of analysis.
Try looking at the strain plots and see if they are true. Displacement is the result of all
the strains, strain plots show how each node expanded. In your case it might be easier
to look at the strain to understand the truth.