The implementation of sensitivity weighting for isogeometric shape optimization on trimmed NURBS-based B-Rep models is realized in this thesis. It is based on isogeometric shape optimization of untrimmed models and isogeometric B-Rep analysis (IBRA). The trimmed NURBS-based B-Rep models were created using the Computer Aided Design (CAD) program Rhinoceros 5 (R).
The gradient based optimization method with sensitivity weighting is applied to these models. After various benchmarks were investigated in this thesis, we have drawn the conclusions that the geometric influence on isogeometric shape optimization can be eliminated and more desirable optimal designs can be obtained.
«