Contact2D problem with a wall-soil interaction model

[ziyanh]

Dear STKO

I am trying to replicate a pseudo-static wall-soil experiment with cyclic displacement and trying to get the earth pressures behind the wall using Contact2D material and BeamContact2D element.

I am simulating the first step so that a small strain condition should be on using elastic material but I can not get resonable results. I attach my model here, can you please help me?

Best regards
Ziyan

[kesavapraba]

Hi, Can you please provide some specific input on your issue. It is challenging to understand your specific issue (i.e., the reasonable result that you expect) from a general statement. Thank you.

[ziyanh]

Dear Prabakaran

Thanks for your response!

I am simulating a wall in contact with the backfill soil that is 0.96m high and 1m wide in the 2D plane. The cyclic point displacement is applied on the wall at the height of 0.87m. I use the units kilonewton for the self-weight of the backfill and kilopascal for the material properties of the wall and soil.

My target is to obtain the earth pressure behind the wall, especially the maximum passive earth pressure achieved when the wall reaches the maximum displacement during each cycle. The earth pressure curve should look like the picture 1 from the study of Caristo et al. (2018):

Picture 1.jpeg (39.8 KiB) Viewed 4696 times

I attach an edited model here and I change the elastic modulus of the wall from 1.1×10^7kPa to 1.1×10^6kPa.

I have several problems in my model now:
1) I see some unexpected flexure when applying self-weight to the backfill, please see picture 2:

Picture 2.jpeg (133.03 KiB) Viewed 4696 times

2) And then when the wall is being pushed and pulled, the horizontal earth pressure is much larger than expectation. Even large negative horizontal earth pressure developed on the top of the backfill when the wall reaches the max positive displacement, please see picture 3:

Picture 3.jpeg (122.37 KiB) Viewed 4696 times

3) Also, there is inconsistence in the horizontal earth pressure during the pushing-pulling process as it should not have negative earth pressure around the mid of the wall, please see picture 4 and 5:

Picture 4.jpeg (124.43 KiB) Viewed 4696 times

Picture 5.jpeg (122.26 KiB) Viewed 4696 times

Best wishes
Ziyan

[STKO Team]

Regarding the model:

• The main mistake was that the soil domain was flipped (its Normal vector - local Z - was pointing downward). In 2D analyses, this results in a negative volume, negative jacobian, and so all results are a mess. I simply used the Reverse command in STKO to reverse the direction of the surface normal vector.
• There were also some wrong settings in the anlyses (large tolerance, few time steps, better use NewtonLineSearch with contact, large gap tolerance in contact element, no flag on initial contact option)

Now it seems to work but I don't think you will be able to get those profiles with the boundary conditions you applied. Please double check the paper and make sure the BC are correct

[ziyanh]

Dear STKO

According to your suggestion last time, I made some changes to the model. Now, the wall's movement seems reasonable; however, the soil mass deformation and the earth pressure development behind the wall are still problematic/unrealistic.

First, let me specify again the object we want to simulate so you can better understand. We conducted a series of experimental soil box tests to investigate the passive earth pressure development under cyclic loadings for integral bridge abutments. The test volume of soil is 1000×1000×960 mm^3. A 1000 mm high moveable wall was placed in front of the soil mass with a hinge at the base to simulate the integral abutment (Luo et al., 2023). More detailed information can be referred to in the paper 'Physical and numerical investigation of integral bridge abutment stiffness due to seasonal thermal loading '(Luo et al., 2023).

r1.jpg (95.93 KiB) Viewed 3460 times

An actuator was set at the height of 870mm on the moveable to impose cyclic loading. The following plot shows an example of the input cyclic loading:

r2.jpg (14.3 KiB) Viewed 3460 times

From the experimental evidence, we see that the pressure behind the wall is below 100 kPa (Luo et al., 2023):

r3.jpg (68.33 KiB) Viewed 3460 times

Then, we conducted a similar numerical analysis with OpenSees; we observed the following issues:
1. A very large earth pressure zone is concentrated on the upper part behind the moveable wall. The shape of the earth pressure curve is unrealistic and quite different from that obtained experimentally.

r4.jpg (66.34 KiB) Viewed 3460 times

2. There is negative passive earth pressure around the lower part of the soil behind the moveable wall, which means there is tension around the base hinge. However, it is expected that the soil mass will be totally compressed when the wall is being pushed towards the soil mass.

What do you think I am doing wrong?

Best regards
Ziyan