I modelled a helical pile (with a mass at the head) installed in dry sand undergoing seismic shaking at the base. I am trying to mimic a full-scale test we performed in a shaking table. I am using ABAQUS.
My modelling steps are as follows:
Step1: Initial conditions which consists of applying fixation at the base and lateral sides of the model. I also fixed the empty void in the soil (Where the soil will be placed, I know this does not include the effect of driving the pile). I applied the initial state of stresses using the predefined field in abaqus.
Step2: Geostatic step. I applied gravity loading on all the soil. At the end of this step, I checked the stress states with depth and they are matching my initial calculations and all deformations are at of the order of 10^-7. (should they be smaller?)
Step3: Removed fixations at the void inside and activated the interaction between the pile and soil. Also applied the gravity load of the pile so that it will settle slightly in the soil. Checked stresses here and displacements and they look okay.
Step4: Applied the time history of my earthquake at the base. At the first 3 seconds during shaking the soil didnt show any settling, however during the strong shaking the soil starts to settle and reaches 0.5m displacement downwards at the surface. (look at the attached images)
I cannot think of any reason causing this. Your help is greatly appreciated!
Attached are the input file used for the analysis (took about 5 days to run) and several screenshots of the deformations during the shaking.
Also attached several videos of the run in difference scaling factors.
https://www.youtube.com/watch?v=5lsB4MpMMIM&feature=youtu.be
https://youtu.be/aPknctWkYuE
Even with self compaction we are talking about 0.5 m compaction in a 4.5 m height soil layer which is significant. I understand that I may expect some self compaction due to densification but not 0.5 m.
Ofcourse I compared with the shaking table results, the soil didnt settle at all.
Hi, regarding the small deformations that you have after the static step, they sound OK if your initial conditions (stress) were close to the actual solution (geostatic stresses)... Regarding the huge deformations after the dynamic part. The settlement will depend on the void ratio variation. Therefore, if that value is too big at the begining the soil will experiment large deformations... Also, from the pictures it seams that the soil settle the same Arron's the pile than far away from it. That doesn't seems correct. I would expect less deformation due to the friction between the pile and the soil... The best way to check your soil parameters is making a small column (a 1D problem with just a few elements). Then apply the ground motion and check the settlement.... Hope this helps you. Cheers!!
This is what I am trying to do now actually and unfortunately the results are not matching deepsoil even though the model is so simple and just includes a soil block. I am really lost
For your information, the soil model im using doesnt take void ratio as input (Mohr Columb) so this is not the reason
I am not an expert in the type of model you are running here, but in my experience anytime I have conducted models of soil, geologic, or hydrogeologic systems, I have found it best to extend my boundary conditions to a point well beyond the zone of influence of the specific area I am evaluating. From the video representation, I note that your zone of influence appears to extend very close to, or even beyond, your modeled boundaries. If that is true of your actual model, I suggest you extend your boundary conditions further away from your study area. That may not resolve your problem, but it will at least eliminate the potential for the boundary conditions to heavily impact your results.
The pattern illustrated by your model output certainly suggests that the soils around and below your pile are consolidating during the shaking. If your model does not allow for the soils to consolidate, perhaps you are loosing soil volume across the boundaries.
- Badges
- Science topic
- Similar topics
- Engineering
- Civil Engineering