I’m looking for a way to apply time varying body force or a ground motion in OepenSees PFEM model . I modified the provided python example in OpenSeespy doc (https://openseespydoc.readthedocs.io/en/latest/src/dambreak.html) to model fluid sloshing in a rectangular tank. I tried the code lines bellow but failed to do it.
Can anyone provide some help or an example of a 2D sloshing tank model.
Thanks
# Horiz Acceleration
tsTag = 300
acctimes = np.arange(0,totaltime+dtmax,dtmax)
accvalues = np.sin(acctimes*2*np.pi/1.5)
ops.timeSeries('Path', tsTag, '-values', *accvalues, '-time', *acctimes, '-factor', 9.81)
msTag = 400
ops.pattern('MultipleSupport', msTag)
gmTag = 500
ops.groundMotion(gmTag, 'Series', '-accel', tsTag)
# change th original BC (fixed) to SupportMotion
for nd in ops.getNodeTags('-mesh', wall_tag):
#ops.fix(nd,1,1)
ops.fix(nd,0,1)
ops.imposedSupportMotion(nd, 1, gmTag)
Abdelghani Seghir The groundMotion command in the OpenSees PFEM (Particle Finite Element Method) model allows for the application of time-varying body forces. A time series may be used to define ground motion, which can subsequently be applied to specified nodes using the imposedSupportMotion command.
You define a time series for horizontal acceleration using the timeSeries command and then apply it to a ground motion using the groundMotion command in your code sample. Using the imposedSupportMotion command, the ground motion is then applied to particular nodes. However, it is unclear from the code sample if the time series is appropriately created and whether the imposedSupportMotion command is applied to the relevant nodes.
More information on the model and the specific problem you're attempting to tackle would be helpful. It would also be beneficial to discuss any difficulties or issues that you experienced when executing the code.
In general, you'd need to do the following to construct a 2D sloshing tank model:
- Define the tank's shape and the fluid domain.
- Define the fluid and tank's material characteristics.
- Apply boundary conditions to the tank's top and walls.
- To the fluid domain, apply time-varying body force or ground motion.
- Carry out the analysis.
The OpenSeespy documentation include sample code for a 2D sloshing tank model. More examples may be found on the OpenSees official website or in other research publications that utilize OpenSees to simulate sloshing.
Thank you Qamar
In the OpenSeespy documentation, we can find 4 models for the dam break problem. I'm trying to modify the one without obstacle, the walls are rigid and there is no FSI. (https://openseespydoc.readthedocs.io/en/latesth/src/dambreak.html)
My first attempt is to impose a body force to the fluid domain by changing the value of b1 in the line : 72
eleArgs = ['PFEMElementBubble',rho,mu,b1,b2,thk,kappa]
A constant value worked well but I haven't found a way to make it time dependent.
A second attempt is to impose a support motion, which seems to require defining a structural mesh and solving the FSI problem. The beam model is used in the document by Zhu and Scott (https://peer.berkeley.edu/publications/2019-06).
I'll try this way.
Qamar Ul Islam
Qamar Ul Islam
I tried the FSI model with beam elements for the tank walls. It worked well, but for some cases of mesh size I get fluid element crossing the beam wall as shown in the picture below. Is there any way to prevent this?
- Badges
- Science topic
- Similar topics
- Engineering
- Modeling