A 2012 #paper on design for #3D-printing #manufacturing in metals. Not much about topology optimization AFAICT.
#pdf #paper on “Multiple-Material #Topology-Optimization of Compliant Mechanisms (#mechanical #flexures) Created Via PolyJet #3D-Printing”. It says that multiple materials has the potential to eliminate the need for thin, weak sections. (PolyJet is a process that selectively deposits either of two liquid resin monomers and polymerizes them with UV. One of the resins is an elastomer.) Their results do not seem to justify that level of enthusiasm, as they still contain thin, weak sections, and they don’t seem to have measured the output force. #manufacturing
#pdf #paper with lots of results of #topology-optimization made by #3D-printing. They wrote and published a MATLAB program called TOPSlicer to translate topology-optimization results to useful formats. #manufacturing
#pdf #paper on #topology-optimization for hyperelastic materials. From glancing at the figures I’m not super impressed with their results, though they sure do have a lot of math. Apparently the authors think “Neo-Hookean” is a person, although at least they managed to avoid citing any papers by him.
#metamaterials (“mesostructures”) via #3D-printing in order to achieve customized and graded elasticity. Still, they seem to just be replacing struts with coils, mostly, although there are some other #flexures mentioned. #manufacturing
#pdf dissertation by Benjamin Vayre in French on #topology-optimization for #3D-printing by electron-beam melting, with a bunch of interesting-looking results. Mentions the acronym DFAM, “Design for Additive Manufacturing”. Mentions topology optimization. Proposes the following design process: analysis of specification → initial form generation (either manually or through topopt) → choice of orientations during manufacturing → geometry optimization → validation → maybe start over. Apparently the dude designed a mounting bracket for his dissertation using off-the-shelf software at the University of Grenoble. I guess not every university can afford to have standards. They have a wicked mounting bracket and e-beam melter though.
#pdf #paper on automatically modifying #topology-optimization results for #3D-printing (in this case, FDM), which mysteriously doesn’t seem to have considered incorporating the #manufacturing constraints into the objective function.
This 2008 #paper on “Design for additive manufacturing of cellular structures” has the best diagram I’ve ever seen of the octet truss (aka fcc crystal structure). They’re doing #topology-optimization of #metamaterials using “particle swarm optimization”, which I hadn’t heard of before, but apparently it’s a cooperative variant of genetic algorithms. They’re apparently doing this to optimize “morphing airfoils”, but their final results are not very impressive.
a #PDF of a super widely cited 2011 #paper on #topology-optimization for #3D-printing, showing some results they got with graded density etc.
A 2003 US patent (filed in 2000) for an I-beam with a zigzag web, which it describes as a prefabricated truss. #patents
Halide is a domain-specific language embedded in C++ for image processing, supporting #GPGPU backends as well as #SSE, by two of the Simit authors.
Comment thread on Simit, a domain-specific language embedded in C++ for simulation, specifically for finite element modeling (#FEA). Fredrik Kjolstad presented it at SIGGRAPH.