Random time-varying concentrated loads?

[mihailmi]

Hi,
Is there a way to impose random time-varying concentrated loads on a structure? Something similar to DLOAD user-subroutine maybe?
I want to load a structure with time-varying pressure forces and viscous forces.

Thanks.

 

[xerf]

As a simple idea for random concentrated load,
you can generate a random sequence of numbers between 0.0 and 1.0 (outside ABAQUS, previous to the analysis), according to whatever probabilistic distribution you want,
then put the random sequence in an *AMPLITUDE definition, and attach the *AMPLITUDE definition to a *CLOAD definition.

...and by the way I was thinking that pressure should be distributed, therefore DLOAD subroutine should be ok for such random force generating coding.

 

[mihailmi]

Hi. Yes DLOAD is perfect for pressure but I need to transfer local forces. The solution of random numbers... I don't think it's a solution. I calculate forces due to fluid motion (pressure forces and viscous forces) and I want to load them on the structure during a *dynamic, implicit analysis with an automatic time step. So I don't know what the time will be at every increment and the amplitude curve it is not one that I impose but the fluid does it.
So...? Any other hints?

 

[jcmbl]

mihailmi, if you got a solution to your problem, please share it, as I've experienced nearly the same difficulties.

Thanks,

J

 

[mihailmi]

Yes, I have the solution now. You need Abaqus 6.5 which has a new loading type: "SURFACE TRACTION" and an associated user subroutine UTRACLOAD. You will still not be able to load forces but you can derive traction pressures from them (ABAQUS 6.4 let's you specify only time-varying normal pressures). Hope it will be useful for you too.

 

[jcmbl]

mihailmi,

can you please share your UTRACLOAD subroutine? How do you pass the pressure magnitudes on? I guess you are doing some flutter analysis, but for static aeroelasticity, a DLOAD wouldnt be more appropriate?

Cheers

 

[mihailmi]

Well, I think you are right. For static aeroelasticity (divergence for example) you could use DLOAD for pressure loads. But since I am doing dynamic aeroelasticity (flutter) I am not only interested in pressure loads but also in shear loads which are important too. DLOAD can only loadpressures normal to surface and shear is tangential. This is why I need UTRACLOAD which can apply a pressure following any direction I want (for my case the resultant of pressure forces and shear forces).
I don't think you can find something interesting in my UTRACLOAD routine since it is case study dependent. But t oanswer your question, I am writing from FLUENT the resultant vector of the forementioned loads on each face element at the interface fluid/structure and I am loading them in ABAQUS with UTRACLOAD where you need to supply T_USER (vector direction) and ALPHA (vector's magnitude) for each element loaded with DLOAD (TRVECxNU).
Best regards,

 

[jcmbl]

mihailmi,
actually I was asking for the bits of your subroutine just to have a model for my problem. I've never used subroutines, so it's a bit confusing at the beginning. From what I understood, the DLOAD subroutine, as well as UTRACLOAD, is called for each integration point. That means I have to read in my subroutine the CP distribution previously interpolated at each of the integration point location and pass actually the magnitude of the CP at a particular integr point. Is that right? I'm doing the static aeroelastic study only. Have you done this before?

Cheers

 

[mihailmi]

Yes the subroutines are called for each integration point and yes you have to pass for DLOAD the pressure previously read from somewhere (this is what I do).