TOPOLOGY OPTIMIZATION TO MINIMIZE STIFFNESS
TOPOLOGY OPTIMIZATION TO MINIMIZE STIFFNESS
(OP)
Hi all,
As you know by default, ANSYS's toplogy optimization, maximizes the stiffness of the structur while minimizeing the volume. I was wondering if it is possible to minimize the structur'e stiffness.
i.e., I have a design in which I would like to have maximum stifness in one direction and minimum stifness in the oder direction. I have to define two load steps, one for each direction and use TOCOMP to define the OBJ function for the loads. But how can I tell ANSYS to minimize one function while maximizing the other one?
Thanks,
Mohsen
As you know by default, ANSYS's toplogy optimization, maximizes the stiffness of the structur while minimizeing the volume. I was wondering if it is possible to minimize the structur'e stiffness.
i.e., I have a design in which I would like to have maximum stifness in one direction and minimum stifness in the oder direction. I have to define two load steps, one for each direction and use TOCOMP to define the OBJ function for the loads. But how can I tell ANSYS to minimize one function while maximizing the other one?
Thanks,
Mohsen





RE: TOPOLOGY OPTIMIZATION TO MINIMIZE STIFFNESS
I must disappoint you: Ansys can't do this in a topology optimization. I'm not sure, if such a thing is even possible in any software.
The only chance you have, is to use parameter optimization instead. By defining two different state variables for the displacement of the structure in the two directions (umax_1 and umax_2). Then define a third variable for the objective function: fobj=umax_1/umax_2. Then tell Ansys to minimize the objective function.
O course, this is not a topology optimization and it only works if you know how the structure approximately looks like.
Hope this helps!
Alex
RE: TOPOLOGY OPTIMIZATION TO MINIMIZE STIFFNESS
Have you tried the workbench DesEx extension? I have recently carried out a series of structural beam optimisations looking at both iterative topology optimisation and parametric goal based optimisation.
From what I found I agreen with Alex. The topology optimisation is good to point you in a direction but to really refine a well-defined model based on output parameters you should use DesEx.
The studies I ran worked across ~14 input parameters and 4 output parameters. In this case I was trying to minimise nodal rotations using the statistical functions in VSCFUN command.
All done via command objects from Workbench.
(As a side note I'm currently looking into using the shape optimisation functionality combined with an ekill command routine. If it works out I'll post my results in this forum)
Richard