Tolerance & testCommand

Post Reply
crisjaqueplaza
Posts: 22
Joined: Thu Jan 09, 2025 2:02 am

Tolerance & testCommand

Post by crisjaqueplaza » Thu Jul 10, 2025 12:45 am

Dear STKO,

As part of my master's thesis, I would like to make a sensibility analysis about the analysis setup in a shear wall model. The main attributes that I want to evaluate are the testCommand and tolerance. The modifications that I would like to make are the following ones:

- testCommand: Norm Displacement Increment Test, Energy Increment Test, Relative Norm Displacement Increment Test, Relative Energy Increment Test.

-Tolerance: values of 10^(-5) , 10^(-4) , 5 * 10^(-3) , 10^(-2).

I will not prove Norm Unbalance testCommand because I am using penalty method constraint, so the convergency might be difficult. I would like to know your opinion about the matrix of testCommand and tolerance that I want to combinate; the values of tolerance that I chose are the default values in STKO and of other softwares like Diana and Abaqus. Also, I incorporated a value or 10^(-5) as a lower bound.

As I said, I would like to know your opinion based on your experience: what things could you suggest? is it a good matrix of combinations to prove?

Many thanks in advance!

STKO Team
Posts: 3069
Joined: Tue Oct 29, 2019 8:45 am

Re: Tolerance & testCommand

Post by STKO Team » Thu Jul 24, 2025 10:34 am

Unfortunately, there is no unique answer, because the values that you use as convergence tolerance depend on many factors:

Let's consider a tolerance of 1.0e-5, and a Norm Disp Incr: in this case it's an absolute value, so it depends on the unit system you are using: maybe it's fine for millimeters, but it might be too permissive for meters.

Post Reply