I have modelled reinforced concrete beam with CDM property but not getting convergence .I have tried every method automatic stabilization, nlgeom but same there is a major difference in load deflection curve of different meshing size .Here is the graph of concrete 40mm, 30mm, 20mm size. How can 20mm size curve behave so differently? Plz, help me.
It's the nature of plastic-damage model implemented in Abaqus. If you want to make sure, build a cube and mesh it with different uniform mesh sizes. You will get different responses especially in the softening part. You can argue about your results maybe around the peak and before the softening starts. But in the softening region, your results are not dependable.
Thanks for reply..
But how we can converge results ? How we come to know which curve is accurate ?
That's the problem, As far as I know, you cannot count on your obtained answer after the divergency with this method. If you want to capture post-peak, you cannot use local plasticity models or you'll see mesh sensivity.
One of complexity associated with FE analysis of RC structures is mesh dependency of results in which the numerical results do not converge to a unique solution as the mesh is refined. Mesh dependency typically occurs when a constitutive model based on smeared cracking approach is used to trace the damage progression in material with stress-strain softening response like your case.
to overcome this drawback you need to define the tensile response based on fracture energy as stress- crack opening displacement (see abaqus help) instead of stress-strain or the second option use element size close to characteristics length of material (this is why you got same results for mesh 40 and 30 mm since these two values are close to characteristic length of material on the other hand 20 mm is far and you got different response) BUT if you define tensile response as stress-strain and the element size is close to characteristic length, the results are valid only for this mesh size.
Thanks for your answer ,
Same problem i am getting on define tensile response based on fracture energy ,it is not getting converge . Now i am using 30 mm mesh size to be accurate
But afterwards when bonding the control beam with FRP , i am getting same load deflection curve, How it can possible on adding strengthening material ,it is not showing any strengthening effect.. Plz help me in this matter. Thanks
- Badges
- Science topic