failed to converge problem but the result is reasonable

[mokiao]

Hi, all
Anyone can help me?

I included the zerolength element to model the wood shear wall in 3D.
I have come up with the converge problem, at 10 secs (the total running time is 12 secs) under the earthquake excitation. But after a series of warning of failed to converge, then it tried "Trying Newton with Initial Tangent .."Trying 10 Substeps" and so on, finally the ground motion is done and I get reasonable results. Can I use that result? or there is a serious problem that I need to fix first.

I used the analysis codes like below:
# constraints Transformation
numberer RCM
system UmfPack -lvalueFact 20;
test NormDispIncr 1.0e-4 100 0;
algorithm Newton
integrator Newmark 0.5 0.25
analysis Transient
set Tol 1.0e-6

#Create the analysis parameters
set TmaxAnalysis [expr $nSteps*$dt]
set testTypeDynamic NormDispIncr
set TolDynamic $Tol
set maxNumIterDynamic 1000
set algorithmTypeDynamic ModifiedNewton
set DtAnalysis [expr $dt]
set Nsteps [expr $nSteps]
set ok [analyze $Nsteps $DtAnalysis]; # actually perform analysis; returns ok=0 if analysis was successful

if {$ok != 0} { ; # analysis was not successful.
# --------------------------------------------------------------------------------------------------
# change some analysis parameters to achieve convergence
# performance is slower inside this loop
# Time-controlled analysis
set ok 0;
set controlTime [getTime];
while {$controlTime < $TmaxAnalysis && $ok == 0} {
set controlTime [getTime]
puts "convergence attempts are initialized..."
set ok [analyze 1 $DtAnalysis]
if {$ok != 0} {
puts "Trying Newton with Initial Tangent .."
test NormDispIncr $Tol 1000 0
algorithm Newton -initial
set ok [analyze 1 $DtAnalysis]
test $testTypeDynamic $TolDynamic $maxNumIterDynamic 0
algorithm $algorithmTypeDynamic
}
if {$ok != 0} {
puts "Trying Broyden .."
algorithm Broyden 8
set ok [analyze 1 $DtAnalysis]
algorithm $algorithmTypeDynamic
}
if {$ok != 0} {
puts "Trying NewtonWithLineSearch .."
algorithm NewtonLineSearch .8
set ok [analyze 1 $DtAnalysis]
algorithm $algorithmTypeDynamic
}
if {$ok != 0} {
puts "Trying 10 Substeps"
algorithm $algorithmTypeDynamic
set DtMod [expr $DtAnalysis/10.0]
set ok [analyze 10 $DtMod]
if {$ok != 0} {
puts "Trying Newton with Initial Tangent with 100 Substeps .."
test NormDispIncr $Tol 1000 0
algorithm Newton -initial
set ok [analyze 10 $DtMod]
test $testTypeDynamic $TolDynamic $maxNumIterDynamic 0
algorithm $algorithmTypeDynamic
}
if {$ok != 0} {
puts "Trying Broyden with 100 Substeps .."
algorithm Broyden 8
set ok [analyze 10 $DtMod]
algorithm $algorithmTypeDynamic
}
if {$ok != 0} {
puts "Trying NewtonWithLineSearch with 100 Substeps.."
algorithm NewtonLineSearch .8
set ok [analyze 10 $DtMod]
algorithm $algorithmTypeDynamic
}
}
}
}; # end if ok !0

puts "Ground Motion Done. End Time: [getTime]"

[kostast88]

Just getting an answer, doesn't mean it's a correct answer (or at least the less mistaken).
Convergence problems usually mean bad modelling or insufficient analysis parameters combination(or both). Without a small and representative protion of your model (zero length directions and constraints are the main suspects) we can't really tell.
Also I saw lvalue 20.Check the note on lvalue parameter:
http://opensees.berkeley.edu/wiki/index.php/UmfPack_SOE

[vesna]

From what you are describing it sounds like that at a certain point of analysis the default analysis objects could not find a solution, but trying different algorithms helped and the analysis was finished.