Static Riks Convergence problem, due to large strain increments
Static Riks Convergence problem, due to large strain increments
(OP)
Hi all,
I have a model of a bi-dimensional sheet of hexagonal lattice of graphene, in which I define elastic-plastic material properties, with a boundary condition that constraints displacement in X in the left side of the sheet.
In the right side, i apply a unitary concentrated force on the right-most nodes of the geometry in the x direction.
My objective is to run a static riks, non linear, analysis, in order to get the stress-strain curve (by the incremental force (lpf) - displacement solution that riks method outputs) up to the ultimate tensile strenght of the whole sheet as its own material.
The problem is that the analysis never converges, the error that the riks method gives me is that it needs a lower time increment. I already set up the total increments from 100 to 400, already tried to variate the initial arc length, i have already tried lowering the time increment up to 1E-7, and i dont know what to do...
From the search i've been doing on this problem, some people advised to use hybrid beam elements, to review my plastic properties, but still i don't know how to converge this analysis..
My goal is just to take from abaqus/cae a graph with the force-displacement solution of some elements until the maximum tensile strenght is reached from an incremental load analysis, to characterize the elastic plastic behaviour in tension of this model.
Ill attach my input file... the geometry was done by a python script that gets the node coordinates from a toolkit named 'scikit-nano'. I think the input file has all the things needed.
EDIT:: the error that appears in the message file is 'Time increment required is less than the minumum required'. At the last attempt before the analysis crash, some lelements have large strain increments, and the solution ends up diverging.
Thanks all,
I have a model of a bi-dimensional sheet of hexagonal lattice of graphene, in which I define elastic-plastic material properties, with a boundary condition that constraints displacement in X in the left side of the sheet.
In the right side, i apply a unitary concentrated force on the right-most nodes of the geometry in the x direction.
My objective is to run a static riks, non linear, analysis, in order to get the stress-strain curve (by the incremental force (lpf) - displacement solution that riks method outputs) up to the ultimate tensile strenght of the whole sheet as its own material.
The problem is that the analysis never converges, the error that the riks method gives me is that it needs a lower time increment. I already set up the total increments from 100 to 400, already tried to variate the initial arc length, i have already tried lowering the time increment up to 1E-7, and i dont know what to do...
From the search i've been doing on this problem, some people advised to use hybrid beam elements, to review my plastic properties, but still i don't know how to converge this analysis..
My goal is just to take from abaqus/cae a graph with the force-displacement solution of some elements until the maximum tensile strenght is reached from an incremental load analysis, to characterize the elastic plastic behaviour in tension of this model.
Ill attach my input file... the geometry was done by a python script that gets the node coordinates from a toolkit named 'scikit-nano'. I think the input file has all the things needed.
EDIT:: the error that appears in the message file is 'Time increment required is less than the minumum required'. At the last attempt before the analysis crash, some lelements have large strain increments, and the solution ends up diverging.
Thanks all,





RE: Static Riks Convergence problem, due to large strain increments
And you should also be aware, that only the increasing part of you plasticity curve is used.
RE: Static Riks Convergence problem, due to large strain increments
My analysis must continue until the ultimate stress of the structure as a whole, and not of a critical element, is reached. I wanted to output a curve of the force applied (LPF since the load applied is a unit), vs the displacement of the left side of the sheet, and transform the force in stress, by dividing the LPF by the sheet's width, and the displacement in strain, so i would end up with a 'stress-strain' curve of the sheet as a model of a continuum material.
I guess my stopping criteria is when the sum of the forces applied at the right end reaches a maximum...like the ultimate strenght of the structure has been reached. I don't need the necking part of the stress-strain curve of the structure, i just need the curve until the ultimate tensile strength.
Thanks again the quick and informative answer!