Running Abaqus from Matlab

[flinner19766]

I want to run an explicit analysis from Matlab. The analysis contains
a
user subroutine. I have the following line in my Matlab script:

!abaqus job=analysis user="vumat.for" interactive

When the analysis tries to compile the user subroutine it says that
it
cannot open the include file 'vaba_param.inc'. When I run it myself
from a DOS window everything is fine.

I've tried adding various Abaqus and MSDEV directories to the path to
no avail. Where is 'vaba_param.inc' located? Has anybody encountered
this problem before?

Best regards,

Cormac








------------------------ Yahoo! Groups Sponsor --------------------~-->
Has someone you know been affected by illness or disease?
Network for Good is THE place to support health awareness efforts!
http://us.click.yahoo.com/RzSHvD/UOnJAA/79vVAA/PMYolB/TM
--------------------------------------------------------------------~->

[Ana Vila Verde]

It should be at \ABAQUS\6.4-1(or the abaqus version number you have
there)\site.

If you do a search using the file name, you'll find it...

Ana Vila Verde

_________________________________
Ana Celia Araujo Vila Verde
Departamento de Fisica
Universidade do Minho
Campus de Gualtar
4710-057 BRAGA
PORTUGAL

Email: ***@fisica.uminho.pt
Tel. (general): [phone]
Fax.: [phone]
__________________________________



----- Original Message -----
From: "flinner19766" <***@yahoo.com>
To: <***@yahoogroups.com>
Sent: Wednesday, March 30, 2005 1:14 PM
Subject: [ABAQUS] Running Abaqus from Matlab

Post by flinner19766
I want to run an explicit analysis from Matlab. The analysis contains
a
!abaqus job=analysis user="vumat.for" interactive
When the analysis tries to compile the user subroutine it says that
it
cannot open the include file 'vaba_param.inc'. When I run it myself
from a DOS window everything is fine.
I've tried adding various Abaqus and MSDEV directories to the path to
no avail. Where is 'vaba_param.inc' located? Has anybody encountered
this problem before?
Best regards,
Cormac
http://groups.yahoo.com/group/abaqus
Yahoo! Groups Links

------------------------ Yahoo! Groups Sponsor --------------------~-->
In low income neighborhoods, 84% do not own computers.
At Network for Good, help bridge the Digital Divide!
http://us.click.yahoo.com/EA3HyD/3MnJAA/79vVAA/PMYolB/TM
--------------------------------------------------------------------~->

[flinner19766]

I did do a search for the file in the abaqus installation directory
but vaba_param.inc didn't show up. Two files (vaba_param_sp.inc and
vaba_param_dp.inc; single and double precision presumably) were in
the 'site' directory. Out of curiosity I renamed vaba_param_sp.inc as
vaba_param.inc and added that directory to the path but I still get
the same error.

Regards,

Cormac

Post by Ana Vila Verde
It should be at \ABAQUS\6.4-1(or the abaqus version number you have
there)\site.
If you do a search using the file name, you'll find it...
Ana Vila Verde
_________________________________
Ana Celia Araujo Vila Verde
Departamento de Fisica
Universidade do Minho
Campus de Gualtar
4710-057 BRAGA
PORTUGAL
Tel. (general): [phone]
Fax.: [phone]
__________________________________
----- Original Message -----
Sent: Wednesday, March 30, 2005 1:14 PM
Subject: [ABAQUS] Running Abaqus from Matlab

Post by flinner19766
I want to run an explicit analysis from Matlab. The analysis

contains

Post by Ana Vila Verde

Post by flinner19766
a
!abaqus job=analysis user="vumat.for" interactive
When the analysis tries to compile the user subroutine it says that
it
cannot open the include file 'vaba_param.inc'. When I run it

myself

Post by Ana Vila Verde

Post by flinner19766
from a DOS window everything is fine.
I've tried adding various Abaqus and MSDEV directories to the path to
no avail. Where is 'vaba_param.inc' located? Has anybody

encountered

Post by Ana Vila Verde

Post by flinner19766
this problem before?
Best regards,
Cormac
http://groups.yahoo.com/group/abaqus
Yahoo! Groups Links

------------------------ Yahoo! Groups Sponsor --------------------~-->
In low income neighborhoods, 84% do not own computers.
At Network for Good, help bridge the Digital Divide!
http://us.click.yahoo.com/EA3HyD/3MnJAA/79vVAA/PMYolB/TM
--------------------------------------------------------------------~->

[Kevin Ding]

Hi, All

I just wonder if someone can help me with running double precision in vumat.
So far, I can only run this subroutine (written in Fortran) in single
precision, say, using real*4. However, when I changed that to real*8, I got
an error message.

Could any of you help me fix that?

Many thanks
Kevin

_________________________________________________________________
Only one more day left to find $5000 on eBay:
http://adfarm.mediaplex.com/ad/ck/705-10129-5668-323?ID=1






------------------------ Yahoo! Groups Sponsor --------------------~-->
What would our lives be like without music, dance, and theater?
Donate or volunteer in the arts today at Network for Good!
http://us.click.yahoo.com/TzSHvD/SOnJAA/79vVAA/PMYolB/TM
--------------------------------------------------------------------~->

[flinner19766]

Kevin:

I tried this before and could not get analyses to run. I believed at
the time that it was something to do with the initial calculations and
checks Abaqus/Explicit carries out on the vumat. It appeared to me that
even when running as double precision Abaqus/Explicit passes data into
the vumat as single precision during these initial checks. As a result,
inside the vumat the data is rubbish.

I couldn't figure out a way around this. I'd be curious to know if the
above is anyway correct.

Best regards,

Cormac

Post by Kevin Ding
Hi, All
I just wonder if someone can help me with running double precision in vumat.
So far, I can only run this subroutine (written in Fortran) in single
precision, say, using real*4. However, when I changed that to real*8, I got
an error message.
Could any of you help me fix that?
Many thanks
Kevin
_________________________________________________________________
http://adfarm.mediaplex.com/ad/ck/705-10129-5668-323?ID=1

------------------------ Yahoo! Groups Sponsor --------------------~-->
What would our lives be like without music, dance, and theater?
Donate or volunteer in the arts today at Network for Good!
http://us.click.yahoo.com/TzSHvD/SOnJAA/79vVAA/PMYolB/TM
--------------------------------------------------------------------~->

[Kevin Ding]

Cormac,

Many thanks for your reply. Since I could not run the subroutine using
something like "real*8" or "double precision", I just tried to run it with
"double" on the command line and write some variables out. But it appeared
that Abaqus/Explicit still used single precision. Does anyone else in this
group have some ideas about this issue?

Your help will be very much appreciated
Kevin

Subject: [ABAQUS] Re: double precision in VUMAT
Date: Thu, 31 Mar 2005 08:56:42 -0000
I tried this before and could not get analyses to run. I believed at
the time that it was something to do with the initial calculations and
checks Abaqus/Explicit carries out on the vumat. It appeared to me that
even when running as double precision Abaqus/Explicit passes data into
the vumat as single precision during these initial checks. As a result,
inside the vumat the data is rubbish.
I couldn't figure out a way around this. I'd be curious to know if the
above is anyway correct.
Best regards,
Cormac

Post by Kevin Ding
Hi, All
I just wonder if someone can help me with running double precision in

vumat.

Post by Kevin Ding
So far, I can only run this subroutine (written in Fortran) in single
precision, say, using real*4. However, when I changed that to real*8,

I got

Post by Kevin Ding
an error message.
Could any of you help me fix that?
Many thanks
Kevin
_________________________________________________________________
http://adfarm.mediaplex.com/ad/ck/705-10129-5668-323?ID=1

_________________________________________________________________
Only one more day left to find $5000 on eBay:
http://adfarm.mediaplex.com/ad/ck/705-10129-5668-323?ID=1






------------------------ Yahoo! Groups Sponsor --------------------~-->
What would our lives be like without music, dance, and theater?
Donate or volunteer in the arts today at Network for Good!
http://us.click.yahoo.com/TzSHvD/SOnJAA/79vVAA/PMYolB/TM
--------------------------------------------------------------------~->

[Kevin Ding]

Just want to let your guys know that I already got the problem fixed by
simply leaving the args from ABAQUS out wen defining the double precision
variables and then using the 'double' on the commend line to make
ABAQUS/Explicit pass double precision data into VUMAT.

Thanks for Cormac's reply.
Kevin

Post by Kevin Ding

Post by flinner19766
I tried this before and could not get analyses to run. I believed at
the time that it was something to do with the initial calculations and
checks Abaqus/Explicit carries out on the vumat. It appeared to me that
even when running as double precision Abaqus/Explicit passes data into
the vumat as single precision during these initial checks. As a result,
inside the vumat the data is rubbish.
I couldn't figure out a way around this. I'd be curious to know if the
above is anyway correct.
Best regards,
Cormac

Post by Kevin Ding
Hi, All
I just wonder if someone can help me with running double precision in

vumat.

Post by Kevin Ding
So far, I can only run this subroutine (written in Fortran) in single
precision, say, using real*4. However, when I changed that to real*8,

I got

Post by Kevin Ding
an error message.
Could any of you help me fix that?
Many thanks
Kevin
_________________________________________________________________
http://adfarm.mediaplex.com/ad/ck/705-10129-5668-323?ID=1

_________________________________________________________________
http://adfarm.mediaplex.com/ad/ck/705-10129-5668-323?ID=1

_________________________________________________________________
SEEK: Now with over 80,000 dream jobs! Click here:
http://ninemsn.seek.com.au?hotmail






------------------------ Yahoo! Groups Sponsor --------------------~-->
Has someone you know been affected by illness or disease?
Network for Good is THE place to support health awareness efforts!
http://us.click.yahoo.com/RzSHvD/UOnJAA/79vVAA/PMYolB/TM
--------------------------------------------------------------------~->

[Kevin Ding]

Hi, All

Being an ABAQUS newbie, I have benefited a lot from this group. Many thanks
for the helps you guys offered.

Just wonder if some of you have done an analysis of sheet metal stamping
using ABAQUS/Explicit. I am currently trying to run the problem mentioned in
the ABAQUS Example Problems Manual (¡§Deep drawing of a square box,¡š
Section 1.5.2) with my VUMAT. This is a 3D problem - however I found only 4
varables pased into my subroutine, ie, ndir=3 and nshr=1. Since I dont have
too much knowledge about the shell element (S4R used in this case) in
ABAQUS. Could anyone give me some ideas about this issue? Should I deal with
this case with a plane strain or a plane stress condition?

Many thanks
Kevin

_________________________________________________________________
SEEK: Now with over 80,000 dream jobs! Click here:
http://ninemsn.seek.com.au?hotmail






------------------------ Yahoo! Groups Sponsor --------------------~-->
Has someone you know been affected by illness or disease?
Network for Good is THE place to support health awareness efforts!
http://us.click.yahoo.com/RzSHvD/UOnJAA/79vVAA/PMYolB/TM
--------------------------------------------------------------------~->

[BenZ]

ndir=3 => 3 normal components for the strain tensor, E11, E22 and
E33.
nshr=1 => 1 tangent components, here E12.
The (1,2,3) is made so that the 1 and 2 basis vectors are in a
tangent plane of the shell, or in the plane of the shell in case of
S4R element. The 3 vector is normal to the element.

In fact this result is not astonishing because I believe the S4R to
be a thin shell, so there is no transverse shearing, so no E23 and
E13. Let me also point out that E33 is computed with the other
components Eij because you do not have any thinning effect in the
classic shell theory, and, moreover, no S33 through the thickness.
E33 is often computed so that the shell is incompressible, this is
how the thicknesses are updated.

BenZ.

Post by Kevin Ding
Hi, All
Being an ABAQUS newbie, I have benefited a lot from this group. Many thanks
for the helps you guys offered.
Just wonder if some of you have done an analysis of sheet metal stamping
using ABAQUS/Explicit. I am currently trying to run the problem mentioned in
the ABAQUS Example Problems Manual (¡§Deep drawing of a square

box,¡¨

Post by Kevin Ding
Section 1.5.2) with my VUMAT. This is a 3D problem - however I

found only 4

Post by Kevin Ding
varables pased into my subroutine, ie, ndir=3 and nshr=1. Since I dont have
too much knowledge about the shell element (S4R used in this case) in
ABAQUS. Could anyone give me some ideas about this issue? Should I deal with
this case with a plane strain or a plane stress condition?
Many thanks
Kevin
_________________________________________________________________
http://ninemsn.seek.com.au?hotmail

------------------------ Yahoo! Groups Sponsor --------------------~-->
In low income neighborhoods, 84% do not own computers.
At Network for Good, help bridge the Digital Divide!
http://us.click.yahoo.com/EA3HyD/3MnJAA/79vVAA/PMYolB/TM
--------------------------------------------------------------------~->

[Kevin Ding]

Hi, BenZ

Many thanks for your reply. I went through the abaqus manual about the shell
elements for abaqus/explicit and still got some questions regarding to my
subroutine:

1. s4r elemet has 6 degrees of freedom, why does abaqus only pass 4 into
vumat? I mean, E23 and E31 should go with the other 4 components, even if
they are zero, to make a 3D problem. My subroutine is for 3D problems.
Should I modify it to a plane strain or a plane stress problem as it
appeared to me that shell elements are sort of plane stress problem (S33=0).

2. In your email, you mentioned s4r is a thin element and no transverse
shearing. But I could run the subroutine without using *transverse shear
stiffness. Are they the same thing?

3. I knew I need to define 'strainInc(*,3) myself to take the thinning into
account, given that the shell is incompressible. Could you give me some some
details or let me know somewhere I can find the useful information.

Hope it will not disturb your daily work flow unduly.

Again, your help is very much appreciated and other people's input is
greatly welcome.
Kevin

Subject: [ABAQUS] Re: DOF in deep cup drawing
Date: Tue, 05 Apr 2005 07:21:30 -0000
ndir=3 => 3 normal components for the strain tensor, E11, E22 and
E33.
nshr=1 => 1 tangent components, here E12.
The (1,2,3) is made so that the 1 and 2 basis vectors are in a
tangent plane of the shell, or in the plane of the shell in case of
S4R element. The 3 vector is normal to the element.
In fact this result is not astonishing because I believe the S4R to
be a thin shell, so there is no transverse shearing, so no E23 and
E13. Let me also point out that E33 is computed with the other
components Eij because you do not have any thinning effect in the
classic shell theory, and, moreover, no S33 through the thickness.
E33 is often computed so that the shell is incompressible, this is
how the thicknesses are updated.
BenZ.

_________________________________________________________________
Are you right for each other? Find out with our Love Calculator:
http://fun.mobiledownloads.com.au/191191/index.wl?page=191191text






------------------------ Yahoo! Groups Sponsor --------------------~-->
What would our lives be like without music, dance, and theater?
Donate or volunteer in the arts today at Network for Good!
http://us.click.yahoo.com/TzSHvD/SOnJAA/79vVAA/PMYolB/TM
--------------------------------------------------------------------~->

[BenZ]

Yo !

Hey, S4R element seems to me to be a thin shell, but it may work as
a thick shell under certain conditions, but I do not remember well.
What make me remind that is the *TRANSVERSE SHEAR option that
activate the transverse shearing modes. OK, now back to your
questions :
1/Do not make any confusion between dofs, computed at nodes, and
element's variables, computed at Gauss points. In the shell theory,
there are 5 dofs at each points of the reference plane that describe
the kinematics of the material segment passing through this plane.
The 6th dof is for the artificial rigity linked to the rotation of
the segment along it's own axis. It allows the connection of beams.

1'/Yes, S4R is a plane stress element with bending and, eventually,
transverse shearing.

2/Without *TRANSVERSE STIFFNESS your shell is, for sure, a thin
shell.

3/Taking thinning into account is not a trivial task and i'm not a
specialist of the question, but i've seen something like exponential
laws.

BenZ.

Post by Kevin Ding
Hi, BenZ
Many thanks for your reply. I went through the abaqus manual about the shell
elements for abaqus/explicit and still got some questions

regarding to my

Post by Kevin Ding
1. s4r elemet has 6 degrees of freedom, why does abaqus only pass 4 into
vumat? I mean, E23 and E31 should go with the other 4 components, even if
they are zero, to make a 3D problem. My subroutine is for 3D

problems.

Post by Kevin Ding
Should I modify it to a plane strain or a plane stress problem as it
appeared to me that shell elements are sort of plane stress

problem (S33=0).

Post by Kevin Ding
2. In your email, you mentioned s4r is a thin element and no

transverse

Post by Kevin Ding
shearing. But I could run the subroutine without using *transverse shear
stiffness. Are they the same thing?
3. I knew I need to define 'strainInc(*,3) myself to take the

thinning into

Post by Kevin Ding
account, given that the shell is incompressible. Could you give me some some
details or let me know somewhere I can find the useful information.
Hope it will not disturb your daily work flow unduly.
Again, your help is very much appreciated and other people's input is
greatly welcome.
Kevin

Subject: [ABAQUS] Re: DOF in deep cup drawing
Date: Tue, 05 Apr 2005 07:21:30 -0000
ndir=3 => 3 normal components for the strain tensor, E11, E22 and
E33.
nshr=1 => 1 tangent components, here E12.
The (1,2,3) is made so that the 1 and 2 basis vectors are in a
tangent plane of the shell, or in the plane of the shell in case of
S4R element. The 3 vector is normal to the element.
In fact this result is not astonishing because I believe the S4R to
be a thin shell, so there is no transverse shearing, so no E23 and
E13. Let me also point out that E33 is computed with the other
components Eij because you do not have any thinning effect in the
classic shell theory, and, moreover, no S33 through the thickness.
E33 is often computed so that the shell is incompressible, this is
how the thicknesses are updated.
BenZ.

_________________________________________________________________
http://fun.mobiledownloads.com.au/191191/index.wl?page=191191text

------------------------ Yahoo! Groups Sponsor --------------------~-->
In low income neighborhoods, 84% do not own computers.
At Network for Good, help bridge the Digital Divide!
http://us.click.yahoo.com/EA3HyD/3MnJAA/79vVAA/PMYolB/TM
--------------------------------------------------------------------~->

[Kevin Ding]

Many thanks for your suggestions, BenZ. You already helped me a lot. I just
borrowed the hard copies of abaqus manuals and will go through them more
carefully.

Thanks again
Kevin

Subject: [ABAQUS] Re: DOF in deep cup drawing
Date: Tue, 05 Apr 2005 19:43:53 -0000
Yo !
Hey, S4R element seems to me to be a thin shell, but it may work as
a thick shell under certain conditions, but I do not remember well.
What make me remind that is the *TRANSVERSE SHEAR option that
activate the transverse shearing modes. OK, now back to your
1/Do not make any confusion between dofs, computed at nodes, and
element's variables, computed at Gauss points. In the shell theory,
there are 5 dofs at each points of the reference plane that describe
the kinematics of the material segment passing through this plane.
The 6th dof is for the artificial rigity linked to the rotation of
the segment along it's own axis. It allows the connection of beams.
1'/Yes, S4R is a plane stress element with bending and, eventually,
transverse shearing.
2/Without *TRANSVERSE STIFFNESS your shell is, for sure, a thin
shell.
3/Taking thinning into account is not a trivial task and i'm not a
specialist of the question, but i've seen something like exponential
laws.
BenZ.

Post by Kevin Ding
Hi, BenZ
Many thanks for your reply. I went through the abaqus manual about

the shell

Post by Kevin Ding
elements for abaqus/explicit and still got some questions

regarding to my

Post by Kevin Ding
1. s4r elemet has 6 degrees of freedom, why does abaqus only pass

4 into

Post by Kevin Ding
vumat? I mean, E23 and E31 should go with the other 4 components,

even if

Post by Kevin Ding
they are zero, to make a 3D problem. My subroutine is for 3D

problems.

Post by Kevin Ding
Should I modify it to a plane strain or a plane stress problem as

it

Post by Kevin Ding
appeared to me that shell elements are sort of plane stress

problem (S33=0).

Post by Kevin Ding
2. In your email, you mentioned s4r is a thin element and no

transverse

Post by Kevin Ding
shearing. But I could run the subroutine without using *transverse

shear

Post by Kevin Ding
stiffness. Are they the same thing?
3. I knew I need to define 'strainInc(*,3) myself to take the

thinning into

Post by Kevin Ding
account, given that the shell is incompressible. Could you give me

some some

Post by Kevin Ding
details or let me know somewhere I can find the useful information.
Hope it will not disturb your daily work flow unduly.
Again, your help is very much appreciated and other people's input

is

Post by Kevin Ding
greatly welcome.
Kevin

Subject: [ABAQUS] Re: DOF in deep cup drawing
Date: Tue, 05 Apr 2005 07:21:30 -0000
ndir=3 => 3 normal components for the strain tensor, E11, E22 and
E33.
nshr=1 => 1 tangent components, here E12.
The (1,2,3) is made so that the 1 and 2 basis vectors are in a
tangent plane of the shell, or in the plane of the shell in case

of

Post by Kevin Ding

S4R element. The 3 vector is normal to the element.
In fact this result is not astonishing because I believe the S4R

to

Post by Kevin Ding

be a thin shell, so there is no transverse shearing, so no E23 and
E13. Let me also point out that E33 is computed with the other
components Eij because you do not have any thinning effect in the
classic shell theory, and, moreover, no S33 through the thickness.
E33 is often computed so that the shell is incompressible, this is
how the thicknesses are updated.
BenZ.

_________________________________________________________________
http://fun.mobiledownloads.com.au/191191/index.wl?page=191191text

_________________________________________________________________
Searching for that dream home? Try http://ninemsn.realestate.com.au for
all your property needs.






------------------------ Yahoo! Groups Sponsor --------------------~-->
Has someone you know been affected by illness or disease?
Network for Good is THE place to support health awareness efforts!
http://us.click.yahoo.com/RzSHvD/UOnJAA/79vVAA/PMYolB/TM
--------------------------------------------------------------------~->

[munir nazzal]

Thank you Benz for your quick response; my UMAT is for a Two surface plasticity model basedon bounding surface model. I just using it now to to do simple triaxial test; applying a confinning pressure in the first step and then applying a specified boundary condition on the upper boundary(moving it to a certain dispalcement); for the first step I specified it in the UMAT to be only elastic, in my second step it will elasto-plastic. I know that the stain is caculted by Abaqus using the stiffness matrix provided by the my UMAT ; therefore I checked the stiffness matrix and nothing was wrong in it.


Munir



---------------------------------
Yahoo! Messenger
Show us what our next emoticon should look like. Join the fun.

[Non-text portions of this message have been removed]






------------------------ Yahoo! Groups Sponsor --------------------~-->
In low income neighborhoods, 84% do not own computers.
At Network for Good, help bridge the Digital Divide!
http://us.click.yahoo.com/EA3HyD/3MnJAA/79vVAA/PMYolB/TM
--------------------------------------------------------------------~->

[BenZ]

Ah yes sorry, you work with a UMAT, not a VUMAT ! So, you are right,
the DDSDDE contribute to the computation of the displacements so of
the strain increment.
I would advice you to use a displacement BC instead of a pressure in
the first step, or to check that the pressure isn't too high, or,
finally, that the material properties are in the good units so that
the structure has a "real" stiffness.

BenZ.

Post by munir nazzal
Thank you Benz for your quick response; my UMAT is for a Two

surface plasticity model basedon bounding surface model. I just
using it now to to do simple triaxial test; applying a confinning
pressure in the first step and then applying a specified boundary
condition on the upper boundary(moving it to a certain
dispalcement); for the first step I specified it in the UMAT to be
only elastic, in my second step it will elasto-plastic. I know that
the stain is caculted by Abaqus using the stiffness matrix provided
by the my UMAT ; therefore I checked the stiffness matrix and
nothing was wrong in it.

Post by munir nazzal
Munir
---------------------------------
Yahoo! Messenger
Show us what our next emoticon should look like. Join the fun.
[Non-text portions of this message have been removed]

------------------------ Yahoo! Groups Sponsor --------------------~-->
What would our lives be like without music, dance, and theater?
Donate or volunteer in the arts today at Network for Good!
http://us.click.yahoo.com/TzSHvD/SOnJAA/79vVAA/PMYolB/TM
--------------------------------------------------------------------~->

[Kevin Ding]

Hi, All

Just got some ideas about thickness update in VUMAT for shell elements, i.e.
calculate the strainInc(*,3), which should be done by the user. The method I
found in ABAQUS manual is:

DE33=-v(DE11+DE22)/(1-v) where v is Poisson's ratio.

I post this just in case someone will need this in the furture.

Cheers
Kevin

_________________________________________________________________
$60,000 prize pool to be won. Three winners. Apply now!
http://www.healthe.com.au/competition.do






------------------------ Yahoo! Groups Sponsor --------------------~-->
What would our lives be like without music, dance, and theater?
Donate or volunteer in the arts today at Network for Good!
http://us.click.yahoo.com/TzSHvD/SOnJAA/79vVAA/PMYolB/TM
--------------------------------------------------------------------~->

[Mr. Tropical]

Hi Cormac,
Why running ABAQUS from Matlab? It's very
interesting..

Post by flinner19766
I want to run an explicit analysis from Matlab. The
analysis contains
a
user subroutine. I have the following line in my
!abaqus job=analysis user="vumat.for" interactive
When the analysis tries to compile the user
subroutine it says that
it
cannot open the include file 'vaba_param.inc'. When
I run it myself
from a DOS window everything is fine.
I've tried adding various Abaqus and MSDEV
directories to the path to
no avail. Where is 'vaba_param.inc' located? Has
anybody encountered
this problem before?
Best regards,
Cormac

__________________________________
Do you Yahoo!?
Make Yahoo! your home page
http://www.yahoo.com/r/hs





------------------------ Yahoo! Groups Sponsor --------------------~-->
What would our lives be like without music, dance, and theater?
Donate or volunteer in the arts today at Network for Good!
http://us.click.yahoo.com/TzSHvD/SOnJAA/79vVAA/PMYolB/TM
--------------------------------------------------------------------~->

[flinner19766]

Hi Mr Tropical:

Basically I want to optimise the material parameters I have in an FEA
model using the curve-fitting features in Matlab such as fminsearch.

Best regards

Cormac

Post by Mr. Tropical
Hi Cormac,
Why running ABAQUS from Matlab? It's very
interesting..

Post by flinner19766
I want to run an explicit analysis from Matlab. The
analysis contains
a
user subroutine. I have the following line in my
!abaqus job=analysis user="vumat.for" interactive
When the analysis tries to compile the user
subroutine it says that
it
cannot open the include file 'vaba_param.inc'. When
I run it myself
from a DOS window everything is fine.
I've tried adding various Abaqus and MSDEV
directories to the path to
no avail. Where is 'vaba_param.inc' located? Has
anybody encountered
this problem before?
Best regards,
Cormac

__________________________________
Do you Yahoo!?
Make Yahoo! your home page
http://www.yahoo.com/r/hs

------------------------ Yahoo! Groups Sponsor --------------------~-->
What would our lives be like without music, dance, and theater?
Donate or volunteer in the arts today at Network for Good!
http://us.click.yahoo.com/TzSHvD/SOnJAA/79vVAA/PMYolB/TM
--------------------------------------------------------------------~->

[Narcis]

Cormac,

Try to run the optimization udner linux machines, not DOS. I am doing the same things as you through Exceed , without any user subroutines and everything is fine.

Narcis


flinner19766 <***@yahoo.com> wrote:


Hi Mr Tropical:

Basically I want to optimise the material parameters I have in an FEA
model using the curve-fitting features in Matlab such as fminsearch.

Best regards

Cormac

Post by Mr. Tropical
Hi Cormac,
Why running ABAQUS from Matlab? It's very
interesting..

Post by flinner19766
I want to run an explicit analysis from Matlab. The
analysis contains
a
user subroutine. I have the following line in my
!abaqus job=analysis user="vumat.for" interactive
When the analysis tries to compile the user
subroutine it says that
it
cannot open the include file 'vaba_param.inc'. When
I run it myself
from a DOS window everything is fine.
I've tried adding various Abaqus and MSDEV
directories to the path to
no avail. Where is 'vaba_param.inc' located? Has
anybody encountered
this problem before?
Best regards,
Cormac

__________________________________
Do you Yahoo!?
Make Yahoo! your home page
http://www.yahoo.com/r/hs

[flinner19766]

Thanks for the suggestion Narcis. However it turned out that I had
not set the working directory correctly in the Matlab script. All
works okay now.

Regards,
Cormac

Post by Mr. Tropical
Cormac,
Try to run the optimization udner linux machines, not DOS. I am

doing the same things as you through Exceed , without any user
subroutines and everything is fine.

Post by Mr. Tropical
Narcis
Basically I want to optimise the material parameters I have in an FEA
model using the curve-fitting features in Matlab such as fminsearch.
Best regards
Cormac

Post by Mr. Tropical
Hi Cormac,
Why running ABAQUS from Matlab? It's very
interesting..

Post by flinner19766
I want to run an explicit analysis from Matlab. The
analysis contains
a
user subroutine. I have the following line in my
!abaqus job=analysis user="vumat.for" interactive
When the analysis tries to compile the user
subroutine it says that
it
cannot open the include file 'vaba_param.inc'. When
I run it myself
from a DOS window everything is fine.
I've tried adding various Abaqus and MSDEV
directories to the path to
no avail. Where is 'vaba_param.inc' located? Has
anybody encountered
this problem before?
Best regards,
Cormac

------------------------ Yahoo! Groups Sponsor --------------------~-->
In low income neighborhoods, 84% do not own computers.
At Network for Good, help bridge the Digital Divide!
http://us.click.yahoo.com/EA3HyD/3MnJAA/79vVAA/PMYolB/TM
--------------------------------------------------------------------~->