Dear friends!
I'm validating a liquefaction problem by FLAC using Finn-Byrne model. But, there is a problem. The recorded pore pressures near the boundaries have large drops while it is not observed in experiment. In the experiment, rigid boundaries were used. There were no absorbing boundaries. Like the experiment I used rigid boundaries. Then I changed the boundary condition to free field and the drops in pore pressure records were decreased (but i have yet the problem) but actually I cannot do this. Can anyone say me solution?
Dear Amin
Was quiet boundaries your initial boundary for the rigid container? You may need to check some of your input parameters before the dynamics simulations. Best of luck
Ademola
Quiet boundaries were not applied because the test was done inside a rigid container. About the input parameters can you explain more accurately?
I'm using the same rigid container for my model but applied dampers at the transient boundaries of the container to enable me use the quiet boundaries. I didn't get similar problems although I'm using P2P not Finn model.
Adeloma,
I'm using flac 2D. I should say that I examined another model such as PM4Sand, but i had the same problem. Please note that the large drops are seen at the zones near the boundary not at the mid zones. In addition, I applied also free field condition to the boundaries, but it had minor effect.
Are the zones along the boundaries get distorted severely? If so, the free-field option in FLAC uses small strain formulation whereas the liquefaction analyses are generally supposed to be performed in large strain mode. You have to provide higher strength and stiffness to the zones along the lateral boundaries. Refer the manual example "pile supported wharf". You can also try compliant base boundary as an alternative to rigid boundary by converting the acceleration into shear stress time history. May I know what was your experiment?
Hi Dinesh
The boundary zones didn't get distorted as you say. I will try shear stress history instead of acceleration and will say you about that. About the side boundaries! should I apply another condition? In the experiment we have a rigid boundary. The experiment is centrifuge modeling of a caisson foundation.
Amin
I think you can simulate your model using rigid boundaries itself. apply X acc on all nodes along base boundary and as well as on the nodes along both the vertical boundaries. free Y on side boundaries. Normally the side boundaries are impermeable and i hope you didn't alter it. This should work. EPP drops can either be due to dilation in soil or seepage.
When using stress boundary conditions instead of acceleration, "apply sxy, apply xquite and apply yquiet " are applied to the bottom. the side boundaries are rigid. Should i use "apply nsress and apply xquite abd apply yquite" to the side boundaries?
Hi Amin,
I suggest you to go for rigid boundary condition with the boundary conditions that I have mentioned in previous reply. Because, stress boundaries will add on to the complexities. Free field boundaries (apply ff) for vertical sides are used when the stress boundary is applied at the base.
- Badges
- Science topic
- Similar topics
- Philosophy
- Moral Philosophy
- Free Will