problem with pile installation ?

[reza378salami]

Hi All,
I’m trying to model a 2D pile and compare with ALP. The first step is soil gravity analysis which is done. The second step is pile installation by using beam-column and connecting the pile nodes (3 DOF) with soil nodes (2 DOF). It can be done by connection beam- column element,(also equalDOF command is used to connect the beam column node to the soil node). Then we have static analysis for superstructure weight. Final step can be ground motion at the base or static pushover. The problem here is: at the second step it fails to converge. If someone has an idea I will be thankful for the help.

wipe
model basicBuilder -ndm 2 -ndf 3

#soil nodes
node 1 0.0 0.0
node 2 2.0 0.0
node 3 3.0 0.0
node 4 4.0 0.0
node 5 6.0 0.0
node 6 6.0 2.0
node 7 4.0 2.0
node 8 3.0 2.0
node 9 2.0 2.0
node 10 0.0 2.0
node 11 0.0 4.0
node 12 2.0 4.0
node 13 3.0 4.0
node 14 4.0 4.0
node 15 6.0 4.0


fix 1 1 1 0
fix 2 1 1 0
fix 3 1 1 0
fix 4 1 1 0
fix 5 1 1 0

equalDOF 6 10 1 2
equalDOF 11 15 1 2

puts "node and fixities done"

#material
nDMaterial PressureDependMultiYield02 1 2 1.8 9.6e3 2.7e4 36 0.1 \
101.0 0.0 26 0.067 0.23 0.06 \
0.27 20 5.0 3.0 1.0 \
0.0 0.77 0.9 0.02 0.7 101.0
set thick 1.0
set Wx 0.0
set Wy [expr -9.81*1.8]
set Bulk 6.3e6

puts "material done"

element quadUP 1 1 2 9 10 $thick 1 $Bulk 1.0 1.0 1.0 $Wx $Wy
element quadUP 2 2 3 8 9 $thick 1 $Bulk 1.0 1.0 1.0 $Wx $Wy
element quadUP 3 3 4 7 8 $thick 1 $Bulk 1.0 1.0 1.0 $Wx $Wy
element quadUP 4 4 5 6 7 $thick 1 $Bulk 1.0 1.0 1.0 $Wx $Wy
element quadUP 5 10 9 12 11 $thick 1 $Bulk 1.0 1.0 1.0 $Wx $Wy
element quadUP 6 9 8 13 12 $thick 1 $Bulk 1.0 1.0 1.0 $Wx $Wy
element quadUP 7 8 7 14 13 $thick 1 $Bulk 1.0 1.0 1.0 $Wx $Wy
element quadUP 8 7 6 15 14 $thick 1 $Bulk 1.0 1.0 1.0 $Wx $Wy

puts "soil elements done"

#gravity recorder
recorder Node -file Gdisplacement.out -time -nodeRange 1 15 -dof 1 2 3 disp


updateMaterialStage -material 1 -stage 0

system BandGeneral
test NormDispIncr 0.01 50 1
constraints Penalty 1.e12 1.e12
integrator Newmark 1.5 1
algorithm Newton
numberer RCM
analysis Transient

analyze 2 5.e0

updateMaterialStage -material 1 -stage 1

analyze 2 5.e1

#refresh analysis
wipeAnalysis
remove recorders

puts "---------------------------------------------------------- "

#adding pile to teh soil
model basicBuilder -ndm 2 -ndf 3

node 3001 3.0 2.0
node 3002 3.0 4.0
node 3003 3.0 6.0

#conection nodes
node 1001 2.0 2.0
node 1002 4.0 2.0
node 1003 2.0 4.0
node 1004 4.0 4.0

equalDOF 1001 9 1 2
equalDOF 1002 7 1 2
equalDOF 1003 12 1 2
equalDOF 1004 14 1 2
equalDOF 3001 8 1 2

set Bpile 0.7
set Hpile [expr $Bpile/2]
set pileArea [expr ($Bpile*$Hpile)]
set Ec 3.e7
set Iy [expr $Hpile*pow($Bpile,3)/12]
set pileTag 10
geomTransf PDelta $pileTag

element elasticBeamColumn 101 3001 3002 $pileArea $Ec $Iy $pileTag
element elasticBeamColumn 102 3002 3003 $pileArea $Ec $Iy $pileTag

set connectArea [expr $Bpile*$Bpile]
set Econnect $Ec
set Iconnect [expr pow($Bpile,4)/12]
set connectTag 11
geomTransf Linear $connectTag

element elasticBeamColumn 201 1001 3001 $connectArea $Econnect $Iconnect $connectTag
element elasticBeamColumn 202 1002 3001 $connectArea $Econnect $Iconnect $connectTag
element elasticBeamColumn 203 1003 3002 $connectArea $Econnect $Iconnect $connectTag
element elasticBeamColumn 204 1004 3002 $connectArea $Econnect $Iconnect $connectTag


set weight 4
set loadStep 10
mass 3003 [expr $weight/9.81] 0.0 0.0
set load [expr -$weight]

#load
pattern Plain 200 "Linear" {
load 3003 0.0 $load 0.0
}

integrator LoadControl [expr 1.0/$loadStep]
numberer RCM
#numberer Plain
#constraints Penalty 1.e12 1.e12
#constraints Plain
constraints Transformation
test NormDispIncr 0.001 15 1
#algorithm Newton
algorithm ModifiedNewton
#algorithm KrylovNewton
system BandGeneral
#system SparseGeneral
#system ProfileSPD

analysis Static

analyze $loadStep
wipe

[vesna]

why don't you do the analysis in the loop?

here is a good example: http://opensees.berkeley.edu/wiki/index ... e_Analysis

[reza378salami]

Thank you Vesna, I will do the analysis in loop.
Is there any reference for knowing the analysis procedure and the commands better.

[vesna]

some of it is explained in Getting Started manual: http://opensees.berkeley.edu/wiki/index ... d_Analysis

some of it is given in the command manual: http://opensees.berkeley.edu/wiki/index ... s_Commands

In general this is a good reference: Cook, R.D., Malkus, D.S., Plesha, M. E., and Witt, R. J., "Concepts and Applications of Finite Element Analysis," 4th edition, John Wiley and Sons publishers, 2002.

[reza378salami]

Dear Vesna,
thank you for your helpful response,
I wrote the code by FluidSolidPorousMaterial with 3D bbarBrick element (each soil nodes with 3DOF) , it works good. But when I used solid-fluid fully coupled material with BrickUP elements (soil nodes with 4DOF) there is a problem and it does not converge, the question is: how can I be sure that the problem is the modelling procedure or analysis part?
Thanks for your patience
reza

[jinlu]

The 4 connection nodes are NOT connected to anything. They're flying in the model...

[reza378salami]

thank you jinlu,
I used the equalDOF to join the connection element to the soil nodes, is it wrong?
soil nodes are slaved to pile connection nodes in x & y
equalDOF 1001 9 1 2
equalDOF 1002 7 1 2
equalDOF 1003 12 1 2
equalDOF 1004 14 1 2
equalDOF 3001 8 1 2
What is your suggestion to join the pile connection elements to the soil?

another question:
Is it necessary to remove the soil element which is replaced by pile or it doesn’t make big error? Or maybe it’s better to replace it with soft weightless material?