rayleigh $a0 0.0 $a1 0.0
is used, where $a1, the stiffness proportional term is applied to the third column representative of the initial stiffness matrix. I would expect this to give the same result as if I applied the stiffness proportional damping to each element. Thus the following two examples should give the same result, but do not (even if the system remains linear). Have I misinterpreted the "rayleigh" command?
1st Example:
Code: Select all
# Define the model builder
model BasicBuilder -ndm 1 -ndf 1
set g 386.40
set Mj [expr 100.0/ $g]
set W [expr 5.0*$Mj*$g]
set k1 234.9
set k2 220.2
set k3 190.6
set k4 146.2
set k5 87.08
set V1 [expr 0.1451*$W]
set V2 [expr 0.1336*$W]
set V3 [expr 0.1143*$W]
set V4 [expr 0.0872*$W]
set V5 [expr 0.0521*$W]
set a0 0.5609
set a1 0.0036
# Define nodes
node 1 0.0
node 2 0.0 -mass $Mj
node 3 0.0 -mass $Mj
node 4 0.0 -mass $Mj
node 5 0.0 -mass $Mj
node 6 0.0 -mass $Mj
# Define single point constraints
fix 1 1
fix 2 0
fix 3 0
fix 4 0
fix 5 0
fix 6 0
# Define force-deformation relationship for spring
uniaxialMaterial ElasticPP 1 $k1 [expr $V1/$k1]
uniaxialMaterial ElasticPP 2 $k2 [expr $V2/$k2]
uniaxialMaterial ElasticPP 3 $k3 [expr $V3/$k3]
uniaxialMaterial ElasticPP 4 $k4 [expr $V4/$k4]
uniaxialMaterial ElasticPP 5 $k5 [expr $V5/$k5]
# id ndI ndJ mats
element zeroLength 1 1 2 -mat 1 -dir 1
element zeroLength 2 2 3 -mat 2 -dir 1
element zeroLength 3 3 4 -mat 3 -dir 1
element zeroLength 4 4 5 -mat 4 -dir 1
element zeroLength 5 5 6 -mat 5 -dir 1
rayleigh $a0 0.0 $a1 0.0
#Define earthquake excitation
set xacc "Path -filePath motions/elcentro.txt -dt 0.02 -factor $g"
set DT 0.02
set nstep 1559
# tag dir
pattern UniformExcitation 1 1 -accel $xacc
recorder Node -file Node1.out -node 2 -dof 1 disp
recorder Node -file Node5.out -node 6 -dof 1 disp
recorder EnvelopeNode -file EnvelopeNode.out -node 6 -dof 1 disp
recorder Element -file Elements.out -time -ele all force
integrator Newmark 0.5 0.25
test EnergyIncr 1.0e-30 20
algorithm Newton
numberer Plain
constraints Plain
system ProfileSPD
analysis Transient
analyze $nstep $DT
Code: Select all
# Define the model builder
model BasicBuilder -ndm 1 -ndf 1
set g 386.40
set Mj [expr 100.0/ $g]
set W [expr 5.0*$Mj*$g]
set k1 234.9
set k2 220.2
set k3 190.6
set k4 146.2
set k5 87.08
set V1 [expr 0.1451*$W]
set V2 [expr 0.1336*$W]
set V3 [expr 0.1143*$W]
set V4 [expr 0.0872*$W]
set V5 [expr 0.0521*$W]
set a0 0.5609
set a1 0.0036
# Define nodes
node 1 0.0
node 2 0.0 -mass $Mj
node 3 0.0 -mass $Mj
node 4 0.0 -mass $Mj
node 5 0.0 -mass $Mj
node 6 0.0 -mass $Mj
# Define single point constraints
fix 1 1
fix 2 0
fix 3 0
fix 4 0
fix 5 0
fix 6 0
# Define force-deformation relationship for spring
uniaxialMaterial ElasticPP 1 $k1 [expr $V1/$k1]
uniaxialMaterial ElasticPP 2 $k2 [expr $V2/$k2]
uniaxialMaterial ElasticPP 3 $k3 [expr $V3/$k3]
uniaxialMaterial ElasticPP 4 $k4 [expr $V4/$k4]
uniaxialMaterial ElasticPP 5 $k5 [expr $V5/$k5]
uniaxialMaterial Elastic 6 0 [expr $a1*$k1]
uniaxialMaterial Elastic 7 0 [expr $a1*$k2]
uniaxialMaterial Elastic 8 0 [expr $a1*$k3]
uniaxialMaterial Elastic 9 0 [expr $a1*$k4]
uniaxialMaterial Elastic 10 0 [expr $a1*$k5]
# id ndI ndJ mats
element zeroLength 1 1 2 -mat 1 -dir 1
element zeroLength 2 2 3 -mat 2 -dir 1
element zeroLength 3 3 4 -mat 3 -dir 1
element zeroLength 4 4 5 -mat 4 -dir 1
element zeroLength 5 5 6 -mat 5 -dir 1
element zeroLength 6 1 2 -mat 6 -dir 1
element zeroLength 7 2 3 -mat 7 -dir 1
element zeroLength 8 3 4 -mat 8 -dir 1
element zeroLength 9 4 5 -mat 9 -dir 1
element zeroLength 10 5 6 -mat 10 -dir 1
rayleigh $a0 0.0 0.0 0.0