Numerical benchmark for topology optimization in 2D

 

cant_bottom cant_bottom cant_bottom

 

 

 

 

 

 

 

Topology optimization algorithms require the following input data: 1) the design domain and its mesh, 2) the service conditions: location and magnitudes, 3) the properties of the material and 5) the optimization model. In specialized literature, different research groups report similar tests. Those tests frequently differ in the entry data. This makes difficult to performs a fair comparison between the results of different algorithms due to differences in the tests. For the purpose of circunvent this issue, we established a benchmark for topology optimization in 2D, considering the similarities and statistics of the most frequently used tests in 103 papers in the specialized literature. All tests are solved with the SIMP method and two more proposed variants: SIMP-SVC and SIMP-SC. Results are reported in the next references.

Full reference

Documents

A CIMAT internal report of the Benchmark and the results of SIMP, SIMP-SVC and SIMP-SC

Thesis of M. Ochoa with the benchmark and topology Optimization algorithms.

The benchmark at Springer (Archives of Computational Methods in Engineering journal)

Source code

SIMP code and numerical benchmark  

 

 

 

 

Centre for Research
in Mathematics (CIMAT) A.C.

CNuMOG

Links related with the CNuMOG

Links