Dear all. I'm getting good results from minimum compliance volume constraint topology optimizations but I'm not getting any good results from optimizations with eigenfrequencies as an objective. Does someone know how to use the 'Eigenfrequency from modal analysis' and 'Eigenfrequency calculated with Kreisselmaier-Steinhauses formula' as an objective in my volume constraint optimizations? And could perhaps upload a working model? This would be appreciated!! For now I'm generating a Frequency step with 10 requested eigenvalues and in the 'edit design respons' tab i'm using this step as a source. I think I got all the settings right. But the results i'm getting are not good. My best result is attached.
A recent paper by Bogdan Bochenek and Katarzyna Tajs-Zielinska entitled
"Minimal compliance topologies for maximal buckling load of columns"
might be of interest in that context.
Instead of the 'Eigenfrequency calculated with Kreisselmaier-Steinhauses formula' as an objective this should be set as either a contraint or a design response. If you have a specific buckling point you need, then use it as a constraint. If it is needed as more of a guideline to make the buckling resistivity as high as possible then make it a design response, just as you would with volume or a factor of safety. I haven't done this type of optimization my self, so I don't know how much help this has been, but moving the equations to different spots of the optimization can generally help find a solution.
A recent paper by Bogdan Bochenek and Katarzyna Tajs-Zielinska entitled
"Minimal compliance topologies for maximal buckling load of columns"
might be of interest in that context.
Thank you all! I have found a solution on the frequency problem. But first concerning the buckling optimization: I'm pretty sure it is not included in the Abaqus Module. At first I tought 'maximizing the fundamental eigenvalue' and 'maximizing the fundamental buckling load' were one and the same. Which it isn't. I get some pretty good result with maximizing the 1st eigenfrequency (eigenvalue). (see attached picture) But only when I include non-structural masses. Thanks again!
- Badges
- Science topic