Yade

aircraft crash test

Asked by jacopo

hi, i need your help. I have to simulate an aircraft crash test on a soil. I already calibrated the soil, now i have to reproduce the aircraft impact. They gave me a nastran file for the aircraft. I converted it in STL. and i am building the simulation. ( i think i ll only need a translation engine to move the aircraft in the horizontal direction, then it will meet the soil and the impact should start). The impact happens at 45 degree respect the soil, then , i rotate the soil of 45 degrees , while the aircraft will move on horizontal direction ( i must follow this procedure). The aircraft is a light aircraft. What i want to ask is:

-> the nastran file contains also the aircraft materials. The stl file gives me only the geometry, it's like only a rigid body. is there a way to import also the material? if not , i think i have to set a general material to the whole aircraft ( but in this way, the model would not reproduce exacly the real aircraft) and then work on contacts untill i reach the "impact deceleration" that i need to reproduce. ( i have the accelerations of CG of the real impact test that they made on laboratory and i have to reproduce it with a DEM model) any suggestion?

thanks for your time.

Question information

Language:
English Edit question
Status:
Answered
For:
Yade Edit question
Assignee:
No assignee Edit question
Last query:
Last reply:
Revision history for this message
Robert Caulk (rcaulk) said : 		#1

Hello,

>>
but in this way, the model would not reproduce exacly the real aircraft
>>

I do not believe there is an easy way to assign "material" properties to different parts of an STL import. However, in my opinion, you've already made bigger simplifications to this "aircraft impact scenario" that will overshadow any improvement in accuracy that might be gained from assigning individual material properties to different parts of the STL import. For example, the STL import will behave as a rigid body no matter what with the current version of Yade and I suspect you are using soil comprised of beachball sized soil particles :-).

Cheers,

Robert

Revision history for this message
jacopo (varrialeee) said : 		#2

Ok Thanks. Now i 'm using The stl file as a text . I mean , i create a file.txt holding all the aircraft facets positions. In this way i could rotate And traslate their positions easily ( The original file.stl/gts had a wrong orientation) . Now i need To give movement To The aircraft. I have To use The traslation engine. I think To create a list of facets, use The translation engine On this list, And set the contacts between soil( spheres bodies) And aircraft.

-> do i have To clump the facets creating a rigid body? Becouse now i don t know if they will stay togheter during the impact .

-> do i have to reconvert My file.txt in a file.gts? In this way you told me the aircraft Will act a rigid body. I would like To see the aircraft deformations, And register the forces acting on the aircraft center of gravity.

-> I know the facet-facet contact doeSn t exist. How can i do?

I know IT S a long project, But i would like To know which Is the best way to follow Becouse i have many ideas about IT, But i don t know if Are possible To do. Thanks for your attention
________________________________
Da: <email address hidden> <email address hidden> per conto di Robert Caulk <email address hidden>
Inviato: giovedì 9 agosto 2018 18:37:39
A: Jacopo Varriale
Oggetto: Re: [Question #671364]: aircraft crash test

Your question #671364 on Yade changed:
https://answers.launchpad.net/yade/+question/671364

    Status: Open => Answered

Robert Caulk proposed the following answer:
Hello,

>>
but in this way, the model would not reproduce exacly the real aircraft
>>

I do not believe there is an easy way to assign "material" properties to
different parts of an STL import. However, in my opinion, you've already
made bigger simplifications to this "aircraft impact scenario" that will
overshadow any improvement in accuracy that might be gained from
assigning individual material properties to different parts of the STL
import. For example, the STL import will behave as a rigid body no
matter what with the current version of Yade and I suspect you are using
soil comprised of beachball sized soil particles :-).

Cheers,

Robert

--
If this answers your question, please go to the following page to let us
know that it is solved:
https://answers.launchpad.net/yade/+question/671364/+confirm?answer_id=0

If you still need help, you can reply to this email or go to the
following page to enter your feedback:
https://answers.launchpad.net/yade/+question/671364

You received this question notification because you asked the question.

Revision history for this message
Jan Stránský (honzik) said : 		#3

Hello,

> do i have To clump the facets creating a rigid body? Becouse now i don t know if they will stay togheter during the impact .

if you use TranslationEngine, the motion of all facets is prescribed. They stay together, but they won't react on the contact with soil. From this point of view, probably using a clump is a better option.

> do i have to reconvert My file.txt in a file.gts?

no, in both ways you just get facets..

> the aircraft Will act a rigid body. I would like To see the aircraft deformations

I think this is most difficult part of your question. Please describe more the experiment (how much is the aircraft crashed etc.)
Currently I feel Yade might not be the right software to use..

> And register the forces acting on the aircraft center of gravity.

this is no problem

> I know the facet-facet contact doeSn t exist. How can i do?

You could implement it, but why do you need facet-facet contact? For modeling aircraft deformations?

cheers
Jan

Revision history for this message
jacopo (varrialeee) said : 		#4

I have To use yade for this project And compare its result with others taken with others software. My aircraft have to crash on a soil with 18 m/S velocity. IT should penetrate The soil around 1/2 meters with the "nose". I would like to see on video the deformations of the aircraft( if something break, like The wings And etc.), that S why i would like To have facet facet contacts. Or, if Is The same, clumping The structure. What do you suggest) . Thanks for the clear about gts -facets. However i don t have To register the deformation , But i would like To see that the aircraft Is taking "damage" during the impact
________________________________
Da: <email address hidden> <email address hidden> per conto di Jan Stránský <email address hidden>
Inviato: lunedì 13 agosto 2018 14:27:33
A: Jacopo Varriale
Oggetto: Re: [Question #671364]: aircraft crash test

Your question #671364 on Yade changed:
https://answers.launchpad.net/yade/+question/671364

    Status: Open => Needs information

Jan Stránský requested more information:
Hello,

> do i have To clump the facets creating a rigid body? Becouse now i don
t know if they will stay togheter during the impact .

if you use TranslationEngine, the motion of all facets is prescribed.
They stay together, but they won't react on the contact with soil. From
this point of view, probably using a clump is a better option.

> do i have to reconvert My file.txt in a file.gts?

no, in both ways you just get facets..

> the aircraft Will act a rigid body. I would like To see the aircraft
deformations

I think this is most difficult part of your question. Please describe more the experiment (how much is the aircraft crashed etc.)
Currently I feel Yade might not be the right software to use..

> And register the forces acting on the aircraft center of gravity.

this is no problem

> I know the facet-facet contact doeSn t exist. How can i do?

You could implement it, but why do you need facet-facet contact? For
modeling aircraft deformations?

cheers
Jan

--
To answer this request for more information, you can either reply to
this email or enter your reply at the following page:
https://answers.launchpad.net/yade/+question/671364

You received this question notification because you asked the question.

Revision history for this message
Jan Stránský (honzik) said : 		#5

Hello,

Instant replay: "Please describe more the experiment (how much is the aircraft crashed etc.)". First you (and we if you want our help) should know what is simulated. I know you wrote 18 m/S velocity and penetrate 1/2 meters, but for not at all experienced person it does not tell much..

> I have To use yade for this project And compare its result with others taken with others software.

ok, makes some sense. Yade **might** (depending on time, dimensions etc.) be a good choice for the soil modeling, but currently definitely not for modeling deformations (especially inelastic) of the aircraft ..
What exactly "its result" should be?
How much time you have for this project? I think it is a critical point. If years, you can try a lot of things, if weeks or months, I would not be very optimistic..
Also the computational power you have might be important..

> I would like to see on video the deformations of the aircraft( if something break, like The wings And etc.), that S why i would like To have facet facet contacts.
> However i don t have To register the deformation , But i would like To see that the aircraft Is taking "damage" during the impact

a few points:
- according to Wikipedia (I don't know STL almost at all): "STL files describe only the surface geometry of a three-dimensional object". This would mean, that even you have facets, they are useless for aircraft modeling even if there would be fatet-facet contact (because it is "surface of surface"). Maybe you could post the stl file on the internet such that we can view it?
- I understand the idea to use facets for modeling the aircraft as a shell. However, you would need some fancy cohesive inelastic facet-facet contact model able to capture shell-like behavior etc... quite a lot of work, implementing, testing..
- all in all, this aircraft modeling you want is not suitable for DEM and Yade..
- To realistically model the crash is complicated even for specialized softwares. I.e., to "see" deformations and damage is no problem, but if you come to velocity, energies...

Maybe you could model the airplane also by spheres, but that would need a lot of preprocessing work..

> Or, if Is The same, clumping The structure. What do you suggest) .

clumping the facets would make it rigid body

My suggestion: I would go step by step. First make the aircraft rigid and simulate the test this much simplified. It might happen that the simulation would take too much time or the soil would react in completely different way you want or........
After you check the simulation is feasible and gives somehow acceptable results, you can make it more complicated.

cheers
Jan

Revision history for this message
jacopo (varrialeee) said : 		#6

hi jan, really thanks for the answer.

The aircraft will crash only with the first part of the fusolage and will penetrate around 1/2 meters inside a soil ( it is built with 18000 spheres body). the crash test should last few seconds. The aircraft will have constant velocity (18 m/s) untill it touches the soil. As i said i don't have to be focused on deformations but i need to register the acceleration of center of gravity of the aircraft.

If i understand , as you suggest me, is better if i consider the aircraft as a rigid body ( so i will not see deformations, is it exacly?), and after this, try to implement a new facet_facet contact in way to be able to see this kind of deformations. Is it right?

->""""""""How much time you have for this project? I think it is a critical point. If years, you can try a lot of things, if weeks or months, I would not be very optimistic..
Also the computational power you have might be important..""""""""

i have few months. The computational time is not a problem becouse i can use a high RAM and CPU computer power (it is a university computer).

->>""""""""" Or, if Is The same, clumping The structure. What do you suggest) .

clumping the facets would make it rigid body

My suggestion: I would go step by step. First make the aircraft rigid and simulate the test this much simplified. It might happen that the simulation would take too much time or the soil would react in completely different way you want or........
After you check the simulation is feasible and gives somehow acceptable results, you can make it more complicated."""""""""""

do you suggest me to clump the facets? in this way i will have a rigid body. Do i have to use sphere-facet contact only ? Becouse honestly i don't even know how to start to implement a facet-facet contact.

I added the STL-file.

________________________________
Da: <email address hidden> <email address hidden> per conto di Jan Stránský <email address hidden>
Inviato: martedì 14 agosto 2018 08:27:53
A: Jacopo Varriale
Oggetto: Re: [Question #671364]: aircraft crash test

Your question #671364 on Yade changed:
https://answers.launchpad.net/yade/+question/671364

    Status: Open => Answered

Jan Stránský proposed the following answer:
Hello,

Instant replay: "Please describe more the experiment (how much is the
aircraft crashed etc.)". First you (and we if you want our help) should
know what is simulated. I know you wrote 18 m/S velocity and penetrate
1/2 meters, but for not at all experienced person it does not tell
much..

> I have To use yade for this project And compare its result with others
taken with others software.

ok, makes some sense. Yade **might** (depending on time, dimensions etc.) be a good choice for the soil modeling, but currently definitely not for modeling deformations (especially inelastic) of the aircraft ..
What exactly "its result" should be?
How much time you have for this project? I think it is a critical point. If years, you can try a lot of things, if weeks or months, I would not be very optimistic..
Also the computational power you have might be important..

> I would like to see on video the deformations of the aircraft( if something break, like The wings And etc.), that S why i would like To have facet facet contacts.
> However i don t have To register the deformation , But i would like To see that the aircraft Is taking "damage" during the impact

a few points:
- according to Wikipedia (I don't know STL almost at all): "STL files describe only the surface geometry of a three-dimensional object". This would mean, that even you have facets, they are useless for aircraft modeling even if there would be fatet-facet contact (because it is "surface of surface"). Maybe you could post the stl file on the internet such that we can view it?
- I understand the idea to use facets for modeling the aircraft as a shell. However, you would need some fancy cohesive inelastic facet-facet contact model able to capture shell-like behavior etc... quite a lot of work, implementing, testing..
- all in all, this aircraft modeling you want is not suitable for DEM and Yade..
- To realistically model the crash is complicated even for specialized softwares. I.e., to "see" deformations and damage is no problem, but if you come to velocity, energies...

Maybe you could model the airplane also by spheres, but that would need
a lot of preprocessing work..

> Or, if Is The same, clumping The structure. What do you suggest) .

clumping the facets would make it rigid body

My suggestion: I would go step by step. First make the aircraft rigid and simulate the test this much simplified. It might happen that the simulation would take too much time or the soil would react in completely different way you want or........
After you check the simulation is feasible and gives somehow acceptable results, you can make it more complicated.

cheers
Jan

--
If this answers your question, please go to the following page to let us
know that it is solved:
https://answers.launchpad.net/yade/+question/671364/+confirm?answer_id=4

If you still need help, you can reply to this email or go to the
following page to enter your feedback:
https://answers.launchpad.net/yade/+question/671364

You received this question notification because you asked the question.

Revision history for this message
jacopo (varrialeee) said : 		#7

let me know if you can see the file i sent you. thanks jan.

________________________________
Da: Jacopo Varriale
Inviato: sabato 18 agosto 2018 21:41:59
A: <email address hidden>
Oggetto: Re: [Question #671364]: aircraft crash test

hi jan, really thanks for the answer.

The aircraft will crash only with the first part of the fusolage and will penetrate around 1/2 meters inside a soil ( it is built with 18000 spheres body). the crash test should last few seconds. The aircraft will have constant velocity (18 m/s) untill it touches the soil. As i said i don't have to be focused on deformations but i need to register the acceleration of center of gravity of the aircraft.

If i understand , as you suggest me, is better if i consider the aircraft as a rigid body ( so i will not see deformations, is it exacly?), and after this, try to implement a new facet_facet contact in way to be able to see this kind of deformations. Is it right?

->""""""""How much time you have for this project? I think it is a critical point. If years, you can try a lot of things, if weeks or months, I would not be very optimistic..
Also the computational power you have might be important..""""""""

i have few months. The computational time is not a problem becouse i can use a high RAM and CPU computer power (it is a university computer).

->>""""""""" Or, if Is The same, clumping The structure. What do you suggest) .

clumping the facets would make it rigid body

My suggestion: I would go step by step. First make the aircraft rigid and simulate the test this much simplified. It might happen that the simulation would take too much time or the soil would react in completely different way you want or........
After you check the simulation is feasible and gives somehow acceptable results, you can make it more complicated."""""""""""

do you suggest me to clump the facets? in this way i will have a rigid body. Do i have to use sphere-facet contact only ? Becouse honestly i don't even know how to start to implement a facet-facet contact.

I added the STL-file.

________________________________
Da: <email address hidden> <email address hidden> per conto di Jan Stránský <email address hidden>
Inviato: martedì 14 agosto 2018 08:27:53
A: Jacopo Varriale
Oggetto: Re: [Question #671364]: aircraft crash test

Your question #671364 on Yade changed:
https://answers.launchpad.net/yade/+question/671364

    Status: Open => Answered

Jan Stránský proposed the following answer:
Hello,

Instant replay: "Please describe more the experiment (how much is the
aircraft crashed etc.)". First you (and we if you want our help) should
know what is simulated. I know you wrote 18 m/S velocity and penetrate
1/2 meters, but for not at all experienced person it does not tell
much..

> I have To use yade for this project And compare its result with others
taken with others software.

ok, makes some sense. Yade **might** (depending on time, dimensions etc.) be a good choice for the soil modeling, but currently definitely not for modeling deformations (especially inelastic) of the aircraft ..
What exactly "its result" should be?
How much time you have for this project? I think it is a critical point. If years, you can try a lot of things, if weeks or months, I would not be very optimistic..
Also the computational power you have might be important..

> I would like to see on video the deformations of the aircraft( if something break, like The wings And etc.), that S why i would like To have facet facet contacts.
> However i don t have To register the deformation , But i would like To see that the aircraft Is taking "damage" during the impact

a few points:
- according to Wikipedia (I don't know STL almost at all): "STL files describe only the surface geometry of a three-dimensional object". This would mean, that even you have facets, they are useless for aircraft modeling even if there would be fatet-facet contact (because it is "surface of surface"). Maybe you could post the stl file on the internet such that we can view it?
- I understand the idea to use facets for modeling the aircraft as a shell. However, you would need some fancy cohesive inelastic facet-facet contact model able to capture shell-like behavior etc... quite a lot of work, implementing, testing..
- all in all, this aircraft modeling you want is not suitable for DEM and Yade..
- To realistically model the crash is complicated even for specialized softwares. I.e., to "see" deformations and damage is no problem, but if you come to velocity, energies...

Maybe you could model the airplane also by spheres, but that would need
a lot of preprocessing work..

> Or, if Is The same, clumping The structure. What do you suggest) .

clumping the facets would make it rigid body

My suggestion: I would go step by step. First make the aircraft rigid and simulate the test this much simplified. It might happen that the simulation would take too much time or the soil would react in completely different way you want or........
After you check the simulation is feasible and gives somehow acceptable results, you can make it more complicated.

cheers
Jan

--
If this answers your question, please go to the following page to let us
know that it is solved:
https://answers.launchpad.net/yade/+question/671364/+confirm?answer_id=4

If you still need help, you can reply to this email or go to the
following page to enter your feedback:
https://answers.launchpad.net/yade/+question/671364

You received this question notification because you asked the question.

Revision history for this message
Jan Stránský (honzik) said : 		#8

> soil ( it is built with 18000 spheres body)

what soil dimensions do you simulate and what size have the soil DEM particles?

> As i said i don't have to be focused on deformations but i need to register the acceleration of center of gravity of the aircraft.

this is possible with rigid body

> If i understand , as you suggest me, is better if i consider the aircraft as a rigid body ( so i will not see deformations, is it exacly?)
> do you suggest me to clump the facets? in this way i will have a rigid body.

Yes, I suggest to consider the aircraft as a rigid body (by clumping the facets) as the first step, it is the easiest configuration to make it run. When this simplified version works, you can always make it more complicated afterwards.
There will be no deformation of the aircraft.

> Do i have to use sphere-facet contact only ?

and of course sphere-sphere contact..

> you suggest ... and after this, try to implement a new facet_facet contact in way to be able to see this kind of deformations. Is it right?
> Becouse honestly i don't even know how to start to implement a facet-facet contact.

Sorry if I was not clear enough. I strongly suggest NOT to implement any facet-facet contact. I just stated that this is one option. The reasons I do not suggest it:
- it would be time consuming
- it would not be easy
- I think it is relatively risky that the approach would not work (the results would be not as expected, the simulation will be too much slowed down...)

> I added the STL-file.
> let me know if you can see the file i sent you. thanks jan.

you cannot put attachments to questions.. please put the file somewhere on the internet and send us the link

cheers
Jan

PS: please try not to include copy of previous answers in your new answers, it makes the thread a bit messy

Can you help with this problem?

Provide an answer of your own, or ask jacopo for more information if necessary.

To post a message you must log in.