Modelling plate with stiffeners
Modelling plate with stiffeners
(OP)
Hi.
I modelled a 500 x 500mm plate with fixed constraints in edges using ABAQUS/Cae. I included two T-beams (50 x 50mm) to the plate as stiffeners. In my first model I used only 4-node shell elements (in both the plate and the stiffeners). In my second model I included the stiffeners as stringers with linear beam elements. When I rendered the shell and beam thicknesses, the models looked similar.
The analysis was elastic static with nonlinear geometric effects turned on. The displacement of the plate middle point of the shell model was twice the displacement of the model with stringers. Therefore it seems that the model with stringers is much stiffer.
Is there any other way to connect beam stiffeners to shell elements in ABAQUS/Cae than the stringer option? I believe that there is something wrong with my stringer usage, does anyone have any ideas of the possible modelling errors (some common pitfalls of stringers)?
-henki
I modelled a 500 x 500mm plate with fixed constraints in edges using ABAQUS/Cae. I included two T-beams (50 x 50mm) to the plate as stiffeners. In my first model I used only 4-node shell elements (in both the plate and the stiffeners). In my second model I included the stiffeners as stringers with linear beam elements. When I rendered the shell and beam thicknesses, the models looked similar.
The analysis was elastic static with nonlinear geometric effects turned on. The displacement of the plate middle point of the shell model was twice the displacement of the model with stringers. Therefore it seems that the model with stringers is much stiffer.
Is there any other way to connect beam stiffeners to shell elements in ABAQUS/Cae than the stringer option? I believe that there is something wrong with my stringer usage, does anyone have any ideas of the possible modelling errors (some common pitfalls of stringers)?
-henki





RE: Modelling plate with stiffeners
-Pure shell model: 11.4mm
-Stiffeners modelled as stringers: 8.7mm
So the difference wasn't that great after all, but still the pure shell model gives approx. 35% higher displacement value. Is this something that would be expected when using beam elements to model the stiffeners?
-henki