Stress constraint composite optimization
Altair EmployeeDear,
During the optimization of our carbon monocoque I get the following error:
*** ERROR # 1800 ***
Stress/strain/force responses are applied to element 1 which
belongs to a topology or free-size domain. For stress-constrained
topology or free-size optimization, please use the STRESS option
on the design variable definition card.
I have done a continuous sizing with only the displacement of a torque as response/constraint to get the correct torsional stiffness. This step was successful and got me a result. I went further with these files to the next step. Namely adding manufacturing constraints and adding an extra static stress response/constraint to prevent failure from some specific loadcases. This gave me error 1800 as mentioned above.
Is this the correct time during the optimization to include the failure criterium? According to the tutorial it is done in this step. Instead of static stress I also tried failure index as a response/constraint with the same error as result. What is the correct method to implement this important constraint?
Kind Regards,
Answers
-
Hi,
Instead of creating a Stress response from response panel, create through Freesize panel.
Since you are using composite material consider using composite stress as response for the monocoque.
0 -
Hi,
But if I have two types of material (hexply and UD) they have both a different max stress they can handle. So with this method I cant say for example that the stress should not exceed 700 Mpa for material 1, and 1560 Mpa for mateiral 2. Is there a method to take both constraints into account in the optimization cycle?
Kind Regards,
0 -
Hi,
This may not be possible during free size optimization.
You can use size optimization method to provide a stress constraint.
0 -
Thank you for your reply, I've tried as you suggested and I still had the same error as result, but isn't the size method necessary to make your model discrete and adding the manufacturing constraints?
In attachment I have my model with the composite failure as response/ constraint (with the error as result), at the moment the control card is free size to size.
I hope you can help me with this problem?
Many thanks,
Jens
0 -
Hi Jens,
Can you remove DesMonocoqueSizee Dsize card?
I am not sure why this is happening but after deleting this I am able to run the analysis. Also you make sure that you have updated all the laminate manufacturing constraints on DCOMP.
0 -
Hi,
Should I update these constraints in the previous design variable (created during free-size)? I created this design variable because I couldn't select the previous one in the free-size panel. When I compare both design variables I noticed following difference. The first one has DCOMP and the other one DSIZE. When I create another variable as a test it appears again as DSIZE. It appears that this is the problem, I've done a check run without any problems. Now the simulation is running, I'll let you know if everything went fine.
To update any manufacturing constraints I'll go to card edit from the existing design variable (since I can't select them through the free-size panel)?
Many thanks for your help
<?xml version="1.0" encoding="UTF-8"?>
<?xml version="1.0" encoding="UTF-8"?>
0 -
Hi,
For sizing you need DCOMP. So update the DCOMP and run the simulation.
0 -
Hi,
You where right, the simulation ran after updating the card. However I get thicknesses which I did not expect. So obviously I did something wrong (I already checked for correct values of material properties, forces and their units). But I was wondering if there is a way to see why the program chooses these thickness and thus violating the constraint of laminate thickness? Is this possible without checking every value entered during the process? It would save me lots of simulation hours to find this problem. Or is there something essential I forgot during this optimisation-step?
Kind Regards,
Jens
<?xml version="1.0" encoding="UTF-8"?>
0 -
I can see that the optimization is not successful, because of the laminate thickness constraint is violated.
May be you should consider changing it.
0
