Jump to content
goncalop

Different Mass on .OUT file and on imported file - Free Size Optimization

Recommended Posts

Hello everyone,

 

I'm currently running an composite optimization on a part and today I noticed that the mass I get on the .out file and on the model I import after the size optimization is different.

 

On this optimization, I try to minimize the compliance of the part while constraining the mass between 3200g and 3600g.

 

Here is what I get on the .out file:

1635471183_outmass.thumb.png.8712cebc5a26dfea6146362c5e9d6361.png

 

Here is what I get on the imported model:

1399479077_HWMASS.thumb.png.305743e8bdce419f81c4c3ebb551eff0.png

 

While on the output file I get a total mass of 3600 as requested, when I import the model, the total mass is now 5322.

 

I've searched around, followed tutorials and nothing helped.

 

I tried opening the *.inc file as suggested on tutorial OS-3400 but the result is the same (the same as if I just import the .fem file resulting from the free size optimization).

 

I've looked on this forum and I found something about the CMF file instead but after the free size optimization, I only get the *.HM.ent.cmf and not the *.HM.comp.cmf as mentioned here:

 

 

I also tried running the optimization with MATINIT set to 1 on opticontrol but once again, the result is the same...

 

Can someone please help?

 

Thank you very much!

Share this post


Link to post
Share on other sites

@Prakash Pagadala,

 

Another thing I just noticed now, when I start running the optimization optistruct creates another property - PCOMPP that doesn't exist on the original model.

 

 

image.thumb.png.865c4c2baa08f2b8926ebd356a97c6fb.png

 

^This the model used for running the free size optimization

 

image.thumb.png.278ffda69160066c164564094758f616.png

 

^and this is the data imported.

 

Also, another thing. When importing the results to start the size optimization, what is the correct way to do it?

 

1) Importing the .fem file to a new model

 

2) Importing the .inc file to the model used to run the free size optimization

 

Thank you very much!

 

Share this post


Link to post
Share on other sites
Guest
2 hours ago, goncalop said:

@Prakash Pagadala,

 

Another thing I just noticed now, when I start running the optimization optistruct creates another property - PCOMPP that doesn't exist on the original model.

 

 

image.thumb.png.865c4c2baa08f2b8926ebd356a97c6fb.png

 

^This the model used for running the free size optimization

 

image.thumb.png.278ffda69160066c164564094758f616.png

 

^and this is the data imported.

 

Also, another thing. When importing the results to start the size optimization, what is the correct way to do it?

 

1) Importing the .fem file to a new model

 

2) Importing the .inc file to the model used to run the free size optimization

 

Thank you very much!

 

Hard to comment without the model. 

Share this post


Link to post
Share on other sites
24 minutes ago, Prakash Pagadala said:

Can you check if the mass before optimization is same as mass at iteration zero?

 

I see MATINIT is not 1 in the out file. 

 

Hello @Prakash Pagadala,

 

The mass is different:

 

Before Optimization with super plies -- 29192

 

Iteration Zero - 3600

 

Yes, I know MATINIT is not 1 on that OUT file I sent you.

 

It is on this one. On this one, the Iteration 0 mass is the same as the model..

 

Despite this, the end result is the same. After free sizing, the masses of the OUT file and the Model are different..

05_9_post_mapping_ready.out

Share this post


Link to post
Share on other sites
26 minutes ago, Prakash Pagadala said:

Hard to comment without the model. 

 

Take a look at your FileTransferLink.

 

I sent you the .hm model.

 

Anything else just ask!

 

Thank you very much!

Share this post


Link to post
Share on other sites

Hello @Prakash Pagadala, how are you?

 

During the weekend I tried different models including the Tutorial OS-3400 and the mass on the .OUT file and after importing is always different (although to different extents -> 1.5X more here and 1.1X on the OS-3400 tutorial).

 

I also never get the *.HM.comp.cmf no matter what I do..

 

If you need anything more let me know!

 

Thank you very much!

Share this post


Link to post
Share on other sites

@Prakash Pagadala,

 

In the meantime I made another model and I no longer have the problem with the new property. I think it might have had something to do with a small problem I had when creating the connectors.

 

I still have the mass problem though..

 

Do you have any news regarding this?

 

Thank you very much!

Share this post


Link to post
Share on other sites

@Prakash Pagadala,

 

Through the afternoon I tried new optimizations with different constraints:

 

09_9 - Just a new mesh, perfectly symmetric with no ctria elements

 

10_9 - The same new mesh but instead of constraining mass, I constrained mass fraction (between 0.11 and 0.123 - this corresponds to 3200 and 3600 according to the original super-ply model mass)

 

10_9_1 - The same new mesh but instead of constraining mass, I constrained volume fraction

 

Unfortunately, the problem persists.

 

Although the mass on the OUT file is around 3600, once imported, I get different results:

 

09_9 - 5605.690

 

10_9 - 5596.463

 

10_9_1 - 5595.804

I've attached the .OUT files so you can take a look. I don't know if they are useful or not...

Sorry for pushing this so much but I think this may be the last problem I'm facing with my thesis and no matter how much I search I can't seem to find a solution for it...

Thank you very much for your help!

09_9_ready.out

10_9_ready.out

10_9_1_ready.out

Share this post


Link to post
Share on other sites
Guest
9 hours ago, goncalop said:

@Prakash Pagadala,

 

Through the afternoon I tried new optimizations with different constraints:

 

09_9 - Just a new mesh, perfectly symmetric with no ctria elements

 

10_9 - The same new mesh but instead of constraining mass, I constrained mass fraction (between 0.11 and 0.123 - this corresponds to 3200 and 3600 according to the original super-ply model mass)

 

10_9_1 - The same new mesh but instead of constraining mass, I constrained volume fraction

 

Unfortunately, the problem persists.

 

Although the mass on the OUT file is around 3600, once imported, I get different results:

 

09_9 - 5605.690

 

10_9 - 5596.463

 

10_9_1 - 5595.804

I've attached the .OUT files so you can take a look. I don't know if they are useful or not...

Sorry for pushing this so much but I think this may be the last problem I'm facing with my thesis and no matter how much I search I can't seem to find a solution for it...

Thank you very much for your help!

09_9_ready.out

10_9_ready.out

10_9_1_ready.out

HI,

 

I have tested on a few models. I observed that the difference is predominant with composite materials. 

 

I am checking the same with experts. I will get back to you soon 

Share this post


Link to post
Share on other sites
Guest

Hi,

 

I got a reply from one of our expert:

the mass you are seeing in the .out file is for a composite model made up of “superplies” which have variable thickness and therefore don’t directly represent a real-life ply, which has constant thickness. If you request OUTPUT, FZTOSZ, then a script is automatically run after the composite freesize optimization which performs some design interpretation on those superplies by slicing them at 4 different Z levels to get discrete ply shapes which have constant thickness. The result is output to the sizing.fem file. So there can easily be some change in mass during that FZTOSZ superply interpretation process. See attached ppt for a more visual explanation of this.

Share this post


Link to post
Share on other sites
So, my goal is to have a deflection of around 1mm at maximum load. At the beginning I was constraining the displacement of some nodes but the optimization was not working. You suggested me to try and constrain volume fraction, mass etc and it worked (https://forum.altair.com/topic/23382-composite-wing-free-size-optimization/?page=2 - here).
 
Through the analysis I was running I was checking that for a weight of +- 3500 I was getting the desired displacement so I started constraining that (mass at 3200 - 3600). I never thought of checking the mass of the model because I had my mind set on cleaning the plies but last week I decided to check the mass of the model and I was very surprised... 
 
My original model has:
 
Mass: 4100g
Max Displacement: +-6mm at maximum pressure
 
My goal is: to reduce the mass (significantly, or at least maintain) while improving the stiffness of the wing.
 
The problem is that with 5000g, I'm not reducing much (although I'm improving significantly the stiffness).
 
Do you have any suggestion?
 
Hope I made myself clear.
 
Thank you very much!

Share this post


Link to post
Share on other sites
Guest

Hi @goncalop 

So the mass after all three phases is not much changed?

 

During size optimization, did you change the responses or retained the same from Freesize?

 

If yes, can you change the responses (to say minimize mass) and check if that helps?

Share this post


Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
You are posting as a guest. If you have an account, please sign in.
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.


×
×
  • Create New...