Hello, I hope you are doing well.
I am modelling a steady state groundwater condition for a simple slope and I am having an error 101, in which the soil body seems to collapse. My intention is to use the pore pressures from this analysis in a subsequent safety calculation to obtain a SSR value.
For this model I defined a head boundary condition in the upper and lower side of the slope, as well as a seepage boundary condition in the slope face (see image above). When performing the calculation I am getting no convergence, with a Mstage multiplier of 0,36 for the phase, and not going forward from that value (see image below).
To solve this issue I decided to change the hydraulic parameters by having a higher permeability soil (I went from 1e-4 m/s to 1e-2 m/s), however my problem was still persistent. Then I decided to increase the "Max unloading steps" in the numerical control parameters (from 5 to 30), and I got little improvement, but not good enough to achieve a convergence. I am curious about this situation because I think this model is not so diificult to solve. As a checking process, I modelled the same situation in SEEP/W in GeoStudio and no issues were found. I also think that an energy head loss of 20 m is too large for the model, and I have checked the magntude of the gradients which also think are not to large.
Is there something I am not considering or seeing for my model? Is it an honest error from the software since I am obtaining the flow from a plastic analysis and a failure must be presented in the soil? Can this be prevented just by a mere change in some of the model inputs or parameters? I have obtained a solution using a drain toe (where I get convergence), however I just want to see this simple situation for the slope being modelled.
Thank you for your time. I hope I made myself clear.
Regards,
JuanP
Dear Juan,
The results show that you have a huge amount of seepage along the slope, about 3/4 of the height of the slope has water flowing out. This leads to a drop of effective stresses close to the slope surface, and since the effective stresses are already low there anyway it will lead to failure of the slope.In reality if you would have water flowing out of the slope over such a significant part of the slope, it would indeed fail.With kind regards,
Dennis Waterman
Dear Dennis,
It is good to realize that a plastic analysis in Plaxis 2D does accoplate the flow analysis with the stresses, which is not the case in SEEP/W where only the flow analysis is computed. I will try to find practical and real solutions to increase the effective stresses.
Thank you so much for your answer.