When doing dynamic anlaysis, I impose a displacement on a mass
and want its reaction forces (very strange, but I have no choice).
I think when doing it we had better use a matrix condensation for
the equivalent stiffness matrix of the dynamic system. And all reaction
forces are calculated from the equivalent stiffness.
For an example:
SDOF: k=10000; m=64; no damping; dt=0.01; d0=0; v0=0; a0=0; ag1=12000. Using Newmark-Beta method(Beta=0.25, Gamma=0.5).
Apply 0.4 displacement on it to calculate the reaction forces.
My solution is:
d1=d0+v0*dt+(0.5-Beta)*dt^2*a0+Beta*dt^2*a1
d1=0.4; therefore a1=16000;
v1=v0+(1-Gamma)*dt*a0+Gamma*dt*a1; therefore v1=80;
Then reaction force on the mass can be calculated by:
R=m*a1+c*v1+k*d1-(-m*ag1)=1796000;
But the result of OpenSees is:
d1=0.4; v1=0; a1=0; R=772000;
I don't know which one is correct? Can some one help me?/
Thanks in advance.
The source file for OpenSees:
model basic -ndm 2 -ndf 3
node 1 0.0 0.0
node 2 0.0 3.0
fix 1 1 1 1
mass 2 64.0 0.0 0.0
geomTransf Linear 1
element elasticBeamColumn 1 1 2 1.0 30000.0 3.0 1
pattern Plain 1 Linear {
sp 2 1 40.0
}
set gaccel "Series -time {0.01} -values {12000.0} -factor 1.0"
pattern UniformExcitation 2 1 -accel $gaccel
recorder Node -file Nodamping.txt -node 2 -dof 1 reaction
#rayleigh 0.2 0.0 0.001 0.0
constraints Transformation
numberer RCM
system SparseGeneral -piv
test EnergyIncr 1.0e-12 100 3
algorithm Newton
integrator Newmark 0.5 0.25
analysis Transient
analyze 1 0.01
Reaction forces during dynamic analysis
Moderators: silvia, selimgunay, Moderators
Frank, thanks for your reply.
It still cannot work. I checked the boundary acceleration and velocity,
and found both are zero. Therefore "reactionIncludingInertia" cannot
work.
I think the critical point is: how to handle "sp" command in dynamic
analysis. I think in the current version of OpenSees, it was handled
statically in dynamic analysis. I don't know if there's one command
to add displacement dynamically as I formulated above?
Thanks
It still cannot work. I checked the boundary acceleration and velocity,
and found both are zero. Therefore "reactionIncludingInertia" cannot
work.
I think the critical point is: how to handle "sp" command in dynamic
analysis. I think in the current version of OpenSees, it was handled
statically in dynamic analysis. I don't know if there's one command
to add displacement dynamically as I formulated above?
Thanks