Altair RISE

A program to recognize and reward our most engaged community members

Nominate Yourself Now!

Compliance unit for Topology Optimization

Hello, I was doing some topological optimization on Altair Inspire and when I compare the results of different model I see that Altair shows a value for compliance. However, this value of compliance has no unit and it seems to be unitless. Do you have any idea why the compliance in Altair is unitless please ? To me compliance is inverse of stiffness thus should be linked to forces and displacements somehow... Thank you for your help !

Find more posts tagged with

English

Inspire

OptiStruct

Accepted answers

All comments

Adriano_Koga

Hi OptiStruct works with consistent units all over the place.

So the units depend on what you're using mm, m, in, ...N, lbf...

Compliance, like you've mentioned is derived from forces and displacements of the structure, as it can be seen in the documentation, and its units depend on your model.

https://2021.help.altair.com/2021/hwsolvers/os/topics/solvers/os/responses.htm#internal_responses_intro_r

Infernox

@Adriano A. Koga
Thank you for your answer. However, if I change the units from millimeters to meters (or even inches), the value of the compliance doesn't change.

Infernox

@Adriano A. Koga
Thank you for your answer. However, if I change the units from millimeters to meters (or even inches), the value of the compliance doesn't change.

The problem is that when I change from SI to Empire system, I see the mass changing but the compliance stays the same. That's why I assume it has to be adimensional. I don't understand why it is adimensional.

Infernox

Hi OptiStruct works with consistent units all over the place.

So the units depend on what you're using mm, m, in, ...N, lbf...

Compliance, like you've mentioned is derived from forces and displacements of the structure, as it can be seen in the documentation, and its units depend on your model.

https://2021.help.altair.com/2021/hwsolvers/os/topics/solvers/os/responses.htm#internal_responses_intro_r

Hello sir,

Sorry to be annoying but I still misunderstand why Altair Inspire is not screening the units of compliance in the result comparison box of topology optimization (maximizing stiffness).
On the figure, you can see that the compliance is unitless and when I change the units from Metric to English for example, the mass is well adapting but the compliance doesn't change. Which means to me that compliance is adimensional which makes no sense to me...

Thank you for your time,

Thomas

PaulAltair

The problem is that when I change from SI to Empire system, I see the mass changing but the compliance stays the same. That's why I assume it has to be adimensional. I don't understand why it is adimensional.

Maybe....

Are you just changing the units system in inspire? Or actually changing the size of your model and the values in your materials? (You need to do both I think if you want to see a change)

e.g. if you are in mm and change inspire to m (or in), I think it is just interpreting the model units in that system (i.e. if your model is 50 units long, it will be 50 mm or m or in as defined) so everything would cancel out... the stiffness would just go from 100N/mm to 100N/m or 100lbf/in and the inverse (compliance) wouldn't change either

You would need to scale your model by 1000 or 25.4 (including gauges if you have shells) or whatever (up or down as appropriate)? And set material values also appropriate to the actual system in use (e.g. modulus in MPa or GPa or imperial units)

Infernox

Maybe....

Are you just changing the units system in inspire? Or actually changing the size of your model and the values in your materials? (You need to do both I think if you want to see a change)

e.g. if you are in mm and change inspire to m (or in), I think it is just interpreting the model units in that system (i.e. if your model is 50 units long, it will be 50 mm or m or in as defined) so everything would cancel out... the stiffness would just go from 100N/mm to 100N/m or 100lbf/in and the inverse (compliance) wouldn't change either

You would need to scale your model by 1000 or 25.4 (including gauges if you have shells) or whatever (up or down as appropriate)? And set material values also appropriate to the actual system in use (e.g. modulus in MPa or GPa or imperial units)

I am changing the units of Inspire. All parameters are changing in term of units and thus values except compliance.

Adriano_Koga

I am changing the units of Inspire. All parameters are changing in term of units and thus values except compliance.

are you opening the .out file from the fun folder to check that?

When Inspire runs the model, in the background, everything is converted to SI (except when you use the 'export' option).

Probably that's why you're always seeing the same value for Compliance in the .out file.

After running the model, it converts back the results (displacements, stresses, ..) to your original units in order to contour your results in the GUI.

PaulAltair

I am changing the units of Inspire. All parameters are changing in term of units and thus values except compliance.

Yes, so, as I said, this is to be expected I think as all values will change relative to each other, you would also need to scale the model to see a change?

Infernox

are you opening the .out file from the fun folder to check that?

When Inspire runs the model, in the background, everything is converted to SI (except when you use the 'export' option).

Probably that's why you're always seeing the same value for Compliance in the .out file.

After running the model, it converts back the results (displacements, stresses, ..) to your original units in order to contour your results in the GUI.

I might not understand but I am not using any .out file. I am just using the viewer of topology optimization results from Altair Inspire. Is that what you call the .out file ?

Infernox

Yes, so, as I said, this is to be expected I think as all values will change relative to each other, you would also need to scale the model to see a change?

Could you please indicate me the procedure you have in mind to see the change ? I don't see how to rescale the model, I am just changing the unit of Altair Inspire through the button on the lower right corner as shown on the figure

Then I am using the TO results comparison tool as shown here :

here you can see the no-change of compliance for deferent unit choice

PaulAltair

Could you please indicate me the procedure you have in mind to see the change ? I don't see how to rescale the model, I am just changing the unit of Altair Inspire through the button on the lower right corner as shown on the figure

Then I am using the TO results comparison tool as shown here :

here you can see the no-change of compliance for deferent unit choice

And, have you checked the size of the parts?

If you need to scale them, you can do that with scale tool: (under geometry)