asdeasoft.net

Hi,

I am modelling a monopile using a combination of beam and continuum elements, embedded in sand, which I have modelled using SanisandMS. The top of the pile is subjected to a monotonic push, and I would like to extract the force required to move the pile as well as its displacement at ground level.

I have attempted to plot the force-displacement curve, but the results did not match my expectations. The results from OpenSees, compared with another finite element software for verification, showed some plastic behaviour. However, after a ratio of \( u_x/D = 0.3 \), the results indicated linear elastic behaviour, which seems odd. After running the analysis in a different software, I found that the results from OpenSees were quite different. Below is the graph showing the results: To extract the reaction force (in Opensees), I retrieved results from the "Reaction Force" in the OpenSees results tab at the node where I applied the specified displacement. I then plotted this against the displacement of the pile node located at ground level.

Below is the model schematics: Could you please advise me on the issue I'm facing? Could it be an error related to the mixture of solid and beam elements I'm using? or maybe my extraction method is wrong? Let me know if you need more details!

On a separate note, I also tested this with a single brick element in a triaxial compression scenario. The force-displacement results are consistent with those from another finite element software. The method for extracting force (reaction force) and displacement is the same as in the monopile example. For this singular element, OpenSees displays results indicating plasticity, and the force-displacement curve reaches a plateau after a certain displacement.

Thanks for your help, and look forward to your reply.

regards
Ryan

It's difficult to tell what the problem is.
The method you used to extract reaction is fine (if you used imposed displacement on top).

To identify the source of the discrepancy, try adding 1 nonlinearity at a time.
Start with only contact, then only plasticity. And see in which one of them you see the difference

HI Admin,

Thank you for your reply. What do you mean by "only contact"?

Extracting results when setting it to a linear elastic material using "updateMaterialStage -material 1 -stage 0 " gives results similar to those of the other FE software.

Thank you for your reply. What do you mean by "only contact"?

I thought you were using both contact (between pile and soil) and soil nonlinear model. Thus I asked to use 1 non-linearity at a time.
If with a linear material the result is the same, you have to find what the difference is in the constitutive model.
Are you comapring 2 softwares using the exact same constitutive model?

Hi,
Sorry for the late reply.

I'm using an equal degree-of-freedom (equalDOF) constraint to create a rigid connection between the pile and the soil, meaning there is no separation between them. I am applying the same constitutive model in both finite element (FE) software programs, and the material parameters are identical as well. However, in OpenSees, I am unable to achieve a reliable curvature in the force-displacement curve; it appears to behave linearly after a certain point. I tried to apply more force at the top of the pile to induce plastic behavior, but the response continues to follow a straight line.

Could you provide any advice? Or perhaps if you have a similar example, that will be great.

Sincerely,
Ryan

Hi STKO,

Is there an update to this query? Your help is appreciated. Thank you!

Unfortunately, without any other details about what the other software is doing, it's really hard to tell where the difference comes from.

Hi,

What other details do you need about the other software? The parameters of the other finite element software have the same inputs as those listed in OpenSees.

I have tried extracting the behaviour when the soil is fully linear elastic, and the results from OpenSees and the other finite element software are the same.

I believe the issue here is that the load-displacement curves extracted from OpenSees do not display any signs of curvature at higher displacements, which is unusual. Instead of indicating some form of plasticity, the curves reflect linear elastic behaviour. As mentioned, I'm using a sanisand model and I have tried with the PDMY02 model as well, I have similar odd results.

Do you happen to have a file that demonstrates a plateauing curve in a force-displacement relationship, preferably one related to soil and structure interactions?

what constitutive model are you using in the other software?

I'm using a SanisandMS soil model by Liu et al. 2019. This is the same model available in OpenSees as well.