Haven't looked at the model file but on the face of it, I'm guessing there might be a problem with either the boundary conditions, the interface or the soil properties.

I recommend first running a linear static analysis to make sure the boundary conditions are correct ( I'm assuming the system will be in equilibrium otherwise this will not work. ) Thereafter check the model with a linear transient analysis ( this will ensure that the mass is correctly assigned and distributed ). Once these are checked, you can go for a nonlinear analysis. 

For more information you can refer this paper : http://web2py.iiit.ac.in/research_centres/publications/download/inproceedings.pdf.9557324ab5686468.445353495f4153594d4d4554524943414c5f4255494c44494e475f54433230375f576f726b73686f705f50617269735f323031335f46696e616c436f70792e706466.pdf

Haven't looked at the model file but on the face of it, I'm guessing there might be a problem with either the boundary conditions, the interface or the soil properties.

I recommend first running a linear static analysis to make sure the boundary conditions are correct ( I'm assuming the system will be in equilibrium otherwise this will not work. ) Thereafter check the model with a linear transient analysis ( this will ensure that the mass is correctly assigned and distributed ). Once these are checked, you can go for a nonlinear analysis. 

For more information you can refer this paper : http://web2py.iiit.ac.in/research_centres/publications/download/inproceedings.pdf.9557324ab5686468.445353495f4153594d4d4554524943414c5f4255494c44494e475f54433230375f576f726b73686f705f50617269735f323031335f46696e616c436f70792e706466.pdf

Hi,

I agree with Vidish Iyer. Try to run your simplified model with linear elastic material behavior under static loads. You may increase model complexity subsequently.

Other points to consider when using nonlinear constitutive equations for the soil are (i) the calculation of the geostatic stress state and (ii) modeling of construction stages. Since soil strength in Drucker-Prager and other models depend on the mean effective stress (i.e. pressure) level, you need the initial stress state in the soil before applying any further loading. I cannot find such solution steps in your APDL file.

Regards

The convergence problem may arise from different sources, improper simulation of the boundary conditions, the amount of the chosen load increment, the selected convergence criteria (strain, stress, displacement, or force criteria). Any as proposed by Vidish Iyer and Daniel Aubram, you may start with linear elastic problem under static load and update the nonlinear sources gradually to trace the errors.  

thanks for reply Vidish Iyer , Daniel Aubram and saad altaan.

i solve the model incase of linear transient analysis and solve well but in case of nonlinear material and soil the program stop solution after 0.13 sec. i can't find where is the error in my model

Hello Safaa

Could you please send me your "ACC" and "ACCZ" file data file please. I would like to check your ANSYS file. It will also be helpful for two of my undergraduate student.

Thank you.

Md. Basir Zisan 

sorry, i didn't see your comment before.

i will send the file data you required and i want to ask you about the hydrodynamic pressure of fluid on the tank wall how can i get it by ANSYS.

Regards

You should use Drucker-Prager model.

Hello all, i am trying to model an underground water tank. How do i mesh both the tank and the surrounding soil?