arbitrary forces at new nodes added for stage construction

[gaziz]

Hi there,

I am trying to simulate a stage construction using 9_4_node_quadUP element following Frank's advice in previous post (http://opensees.berkeley.edu/community/ ... =2&t=62065).

Now when I am adding new nodes and element, I get tensile stresses before even self-weight of that element is in effect.
Checking nodal forces I see that I actually have arbitrary nodal forces in all the nodes related to new element upon its creation. Apparently these forces cause tensile stresses in the element.

Is there anyway to get rid of these forces? The input script is provided below.

P.S. Similar situation is with quadUP elements, so I don't really have choices for 2d UP elements to simulate stage construction.

In the script below two stages are considered.
In Stage 1, nodes 1-9 (with fixities) and element 1 are generated, and a self-weight of element 1 is applied in linear pattern (1000-6000)sec.
In Stage 2, nodes 10-15 (with fixities) and element 2 are generated, and self-weight of element 2 is applied in linear pattern (7000-12000)sec.

INPUT:
###################################################################
wipe
########
#Stage 1#
########
model BasicBuilder -ndm 2 -ndf 3
# create nodes
node 1 0.0 0.0
node 2 1.0 0.0
node 3 1.0 1.0
node 4 0.0 1.0
# boundary conditions
fix 1 1 1 1
fix 2 0 1 1
fix 3 0 0 1
fix 4 1 0 1
model BasicBuilder -ndm 2 -ndf 2
# create nodes
node 5 0.5 0.0
node 6 1.0 0.5
node 7 0.5 1.0
node 8 0.0 0.5
node 9 0.5 0.5
# boundary conditions
fix 5 0 1
fix 8 1 0
# create material
nDMaterial ElasticIsotropic 1 25000 0.35 1.7
# create element
element 9_4_QuadUP 1 1 2 3 4 5 6 7 8 9 1.0 1 5e-6 1.0 1.0 1.0 0.0 -9.81
# create analysis
constraints Plain
test NormDispIncr 1.0e-10 35 1
algorithm Newton
numberer RCM
system SparseSYM
integrator Newmark 0.5 0.25
analysis Transient
# record nodal displacment
eval "recorder Node -file 1DdisplacementStage1.out -time -nodeRange 1 9 -dof 1 2 disp"
eval "recorder Node -file 1DpwpStage1.out -time -nodeRange 1 9 -dof 3 vel"
eval "recorder Element -file 1DforceStage1.out -time -ele 1 force"
# record elemental stress and strain
recorder Element -file 1DstressStage1.out -time -ele 1 material 1 stress
recorder Element -file 1DstrainStage1.out -time -ele 1 material 1 strain
# create self-weight pattern
pattern Plain 1 {Series -time {1000 6000 1.0e10} -values {0 1 1} -factor 1} {
eleLoad -ele 1 -type -selfWeight 1.0 1.0
}
# analyze
for {set numIncr 1} {$numIncr <= 12} {incr numIncr 1} {
puts "##### STAGE 1: Step $numIncr #####";
analyze 1 5.0e2
}
puts "Finished with STAGE 1"

########
#Stage 2#
########
model BasicBuilder -ndm 2 -ndf 3
# create nodes
node 10 1.0 2.0
node 11 0.0 2.0
# boundary conditions
fix 10 0 0 1
fix 11 1 0 1
model BasicBuilder -ndm 2 -ndf 2
# create nodes
node 12 1.0 1.5
node 13 0.5 2.0
node 14 0.0 1.5
node 15 0.5 1.5
# boundary conditions
fix 14 1 0
# create material
nDMaterial ElasticIsotropic 2 25000 0.35 1.7
# create element
element 9_4_QuadUP 2 4 3 10 11 7 12 13 14 15 1.0 2 5e-6 1.0 1.0 1.0 0.0 -9.81
# check and record nodal forces at this instance
set forcesel1 [open nodalForces_el1.out w]
set el1_forces [eleResponse 1 forces]
puts $forcesel1 "$el1_forces"
close $forcesel1
set forcese2 [open nodalForces_el2.out w]
set el2_forces [eleResponse 2 forces]
puts $forcese2 "$el2_forces"
close $forcese2
# record nodal displacment
eval "recorder Node -file 1DdisplacementStage2.out -time -nodeRange 1 15 -dof 1 2 3 disp"
eval "recorder Node -file 1DpwpStage2.out -time -nodeRange 1 15 -dof 3 vel"
eval "recorder Element -file 1DforceStage2.out -time -eleRange 1 2 force"
# record elemental stress and strain
recorder Element -file 1DstressStage2.out -time -eleRange 1 2 material 1 stress
recorder Element -file 1DstrainStage2.out -time -eleRange 1 2 material 1 strain
# create self-weight pattern
pattern Plain 2 {Series -time {7000 12000 1.0e10} -values {0 1 1} -factor 1} {
eleLoad -ele 2 -type -selfWeight 1.0 1.0
}
# analyze
for {set numIncr 1} {$numIncr <= 12} {incr numIncr 1} {
puts "##### STAGE 2: Step $numIncr #####";
analyze 1 5.0e2
}
puts "Finished with STAGE 2"

wipe
###################################################################

Below nodal forces at corner nodes at the moment of adding element 2 are given. Notice that
element 1 and 2 share nodes 3 and 4, the forces at these nodes are not consistent between the two elements.
element1
node1
0.45422352980592723000 3.06376922247634380000 4.90500000000000020000
node2
0.00000000019438194804 2.20079976809378760000 4.90500000000000020000
node3
0.00000000003968139012 -0.00000000019958519559 -4.90500000000000020000
node4
-0.40874592390274928000 -0.00000000026951052501 -4.90500000000000020000
element 2
node 4
-3.80767799946703930000 -4.77011888117276590000 4.90499999999999940000
node3
2.91977397488580510000 -3.27629192357041090000 4.90499999999999940000
node10
-0.99558265064582829000 -0.06159782495557300500 -4.90499999999999940000
node11
1.26027536884065090000 -0.33702903857091537000 -4.90499999999999940000

Hopefully this makes sense. Any advice would be appreciated.

Thank you!

[fmk]

that element does not subtract off any initial displacements from the current displacements and thus is giving this error. if really needed i can modify the source code for the element

[gaziz]

Thank you for reply Frank!

That would be really great! I don't really have options now for UP elements left and really need to simulate construction stages with UP formulation.

P.S. One thing I noticed is if I run the same problems but with all side nodes fixed in x-dir, I don't have any issues. But that is only ideal situation, barely existing in BVP with many elements.

[gaziz]

The code above works well using the latest TAG as it was fixed by Frank McKenna (Many Thanks!)