Weight problem in free size optimization
Altair EmployeeI have a problem with free size optimization.
My weight limit is 35 kg. The optimization runs and converges, results make sense.
When I import the *.fem file in Hypermesh I am checking the weight of the component usint Tools - mass calc. The problem is that I get an higher value (50 kg) of what the constraint is (and in the output file i clearly see that the weight constraint is respected). It is not a question of considering other component in the calculation as I have only one.
I had a look also at the formula student monocoque tutorial and files and I see that if I try to import the file ' 04_freesize_sizing.18.fem' the weight is 57 kg and not 50 as it should be.
Where am I wrong?
Answers
-
Hi Fede_Guru,
The weight limit you have imposed is for design space.
can you cofirm whether the weight you have calculated using Mass calc, is it for only for design space or entire model?
Also, please share an image of Dconstraint for mass.
0 -
Altair Forum User said:
Hi Fede_Guru,
The weight limit you have imposed is for design space.
can you cofirm whether the weight you have calculated using Mass calc, is it for only for design space or entire model?
Also, please share an image of Dconstraint for mass.
I have just added a file in your dropbox folder.
Yes I am aware of that, and in the mass I am checking only the weight for my design domain..
0 -
Thank Fede_Guru,
I will have a look at the model and get back to you soon.
0 -
Hi Fede_Guru,
is it possible to share the model with us?
0 -
Altair Forum User said:
Hi Fede_Guru,
is it possible to share the model with us?
Unfortunately I am working at it at the moment so it would not be ready for solving.
You can try with the file I put in your dropbox folder (from a tutorial): solve it and read the total mass...the constraint is 50 kg, but if after the optimization I import the solver deck and read the mass i get 57...
0 -
Thanks for sharing the file.
I will get back to you soon.
0 -
Hi Fede_Guru,
I feel the results are OK. the Mass constranint which you have is for the design space not to the entire structure (NOTE: total mass you mentioned is the print from the solver not the mass constraint).
So, the design space mass was reduced and is within 50Kgs.
The global response just means subcase independent. Mass or volume is calculated regardless of subcase definition. Displacement or stress would be subcase dependent.
You can define mass on the whole structure or any component.
0 -
Altair Forum User said:
Hi Fede_Guru,
I feel the results are OK. the Mass constranint which you have is for the design space not to the entire structure (NOTE: total mass you mentioned is the print from the solver not the mass constraint).
So, the design space mass was reduced and is within 50Kgs.
The global response just means subcase independent. Mass or volume is calculated regardless of subcase definition. Displacement or stress would be subcase dependent.
You can define mass on the whole structure or any component.
Unfortunately I am still confused.
with reference to the file you have solved, if you go to Tool - mass calc after you import the solver deck you will find out that the mass of the design domain is 20 kg (and the upper bound is 50 kg)
To me it does not make sense that the compliance cannot increase adding material (so i cannot understand, if this is the case, how can be possible that the solution converged).
I still think that the upper bound of 50 kg is the total mass. If you have a look at the attached pdf of the tutorial you will see in the last page that in the first iteration the mass drops from 67 to 50 kg, and 67 is the total mass that you can read in the attached file 02_analysiscomplete.hm.
Thanks a lot for your help, I am looking forward to hearing what's wrong..
0 -
Fede_Guru,
Stiffness is not only on material but also dependent on geometry.
The mass was 67 before changing the ply thicknessess for CFK 0, 45, -45 and 90 to 2.
The TOTAL MASS is the constraint on the design space which is 50Kgs. Here total mass doesn't mean that the entire mass of the structure.
0 -
Altair Forum User said:
Fede_Guru,
Stiffness is not only on material but also dependent on geometry.
The mass was 67 before changing the ply thicknessess for CFK 0, 45, -45 and 90 to 2.
The TOTAL MASS is the constraint on the design space which is 50Kgs. Here total mass doesn't mean that the entire mass of the structure.
Ok, so why 50 kg are compared to 67 in the end of the pdf (which is the total mass of the structure?)
Of course it depends also on the geometry, but it is simply not possible that you cannot increase the stiffness of a structure adding 30 kg and still being within the design constrain.
Anyway I made another clear example: see the attached file ex.
There is only one geometry and the mass constraint is 0.01. After the optimization I import the solver deck and read a mass of 0.016 with Tool-mass calc.
You can try to run the optimization ad see it by yourself.
Thanks again for your help, I hope we can fix this.
0 -
Hi,
In you model, the intial mass is 1.13e-1 which is 0.1134 and mass after optimization is 1.67e-2.
<?xml version="1.0" encoding="UTF-8"?>
0 -
Also,
please look at the plythickness values in your model.
The material is removed where it is not necessary and added where it is needed.
0 -
Altair Forum User said:
Also,
please look at the plythickness values in your model.
The material is removed where it is not necessary and added where it is needed.
Right, but look at the constraint: the mass should not become larger than 0.01 but in the final iteration the mass is 0.016.
Why this?
0
