Earthquake analysis

Forum for OpenSees users to post questions, comments, etc. on the use of the OpenSees interpreter, OpenSees.exe

Moderators: silvia, selimgunay, Moderators

Post Reply
Jhno
Posts: 214
Joined: Sat May 05, 2012 2:55 pm
Location: Laval

Earthquake analysis

Post by Jhno »

Hi,
looking at the example for freefieldsingle analysis (http://opensees.berkeley.edu/wiki/index ... Single.tcl) I see the earthquake excitation is being applied as a force history (a velocity history multiplied by a factor that results in a force).
I was wondering if it is applied this way because there is no density defined in the x direction for soil elements in the column which would result if the earthquake was applied a an acceleration time history in zero intertia forces? Or maybe there is another reason? Thanks!
mcganncr
Posts: 12
Joined: Tue May 27, 2008 4:22 pm
Location: University of Canterbury

Re: Earthquake analysis

Post by mcganncr »

The ground motion is applied in this way to account for an underlying elastic halfspace following the method of Joyner and Chen (1975), which I think is referenced in the wiki article for that example. All of the nodes still have mass, and hence inertia forces, just as would be the case in a fixed-base analysis using the acceleration as input with uniformExcitation. The difference is that the fixed-base approach inherently assumes that the material underlying the analysis domain is infinitely stiff in comparison to the model domain, while the Joyner-Chen approach allows for the consideration of an underlying base material of infinite rigidity.

Cheers,

Chris
Jhno
Posts: 214
Joined: Sat May 05, 2012 2:55 pm
Location: Laval

Re: Earthquake analysis

Post by Jhno »

" All of the nodes still have mass, " : nope, they do not.

In the example (found here : http://opensees.berkeley.edu/wiki/index ... Single.tcl) they define the material in this way :
set wgtX 0.0
set wgtY [expr -9.81*$rho]
for {set j 1} {$j <= $numEleY} {incr j 1} {
set nI [expr 2*$j - 1]
set nJ [expr $nI + 1]
set nK [expr $nI + 3]
set nL [expr $nI + 2]
element quad $j $nI $nJ $nK $nL 1.0 "PlaneStrain" 1 0.0 0.0 $wgtX $wgtY
}
This does apply weight (i.e. force, not mass) in the Y direction, but no in the x direction.
I might be wrong but I do not see anywhere in the script a command "mass $nodeId" so I doubt they assign any mass to the node.

If you run the example and you type in "print node" there is not a single node with a mass.
fmk
Site Admin
Posts: 5884
Joined: Fri Jun 11, 2004 2:33 pm
Location: UC Berkeley
Contact:

Re: Earthquake analysis

Post by fmk »

set rho 1.7

mass in continuum problems typically comes from the element as computing mass terms for the nodes would be a nightmare for all but regular meshes.
Jhno
Posts: 214
Joined: Sat May 05, 2012 2:55 pm
Location: Laval

Re: Earthquake analysis

Post by Jhno »

Thanks for the answer.
Post Reply