I am currently working on a non-linear analysis of a concrete wall using a non-linear material model in RFEM 6. However, I consistently encounter the following error: The stiffness matrix is singular
I have tested several approaches to resolve the issue. I tried releasing the moment around the X-axis, and I also defined member hinges between the steel beam and the concrete wall, including modelling the connection with rotational springs. However, none of these approaches have resolved the problem.
Could you please review the model and help identify the cause of this instability?
For example, the structural modification to account for reinforcement is not activated in your model.
In addition, the concrete must not be temperature-dependent at the moment:
Thank you for your response. I have now updated the model according to your comments, but the program still gives the message: “The calculation did not converge, but the result might be applicable.”
I have tested different values for the load increments (10, 20, and 30) and also increased the number of iterations, but the calculation still does not converge. I also tested the same model using a linear analysis, where the model appeared to be stable. Therefore, I am wondering whether the issue is mainly related to instability in the nonlinear analysis or simply insufficient iterations.
The calculation time is also very long. At the moment, one load combination takes approximately three hours to run. Is there any way to reduce the computation time?