The Effect of Reliability Index Values on Resulting Reliability-Based Topology Optimization Configurations: Numerical Validation by Shape Optimization.

摘要

Introduction. The classical topology optimization leads to structural type and general layoutprediction and gives a rough description of the shape of both the external and internalstructure boundaries. However, Reliability-Based Topology Optimization (RBTO) modelproduces multiple reliability-based topologies with high levels of performance. The aim ofthis work is to study the effect of reliability changes on the obtained topologies.Materials and Methods. The developed Gradient-Based Method (GBM) has been usedefficiently as a general method for several applications (statics and dynamics). When consideringseveral reliability levels, several topologies can be obtained. In order to comparethe resulting topologies, a shape optimization is considered as a detailed design aspect.Results. Numerical applications are carried out on an MBB (Messerschmitt-B?lkowBlohm)beam subjected to a distributed load. The DTO model is carried out without considerationof reliability concept. However, for the RBTO model, an interval of reliability isconsidered that produces several topologies. Here, the randomness is applied on geometryand material parameters. The application of the shape optimization algorithm leads toreduced structural volumes when increasing the reliability levels.Discussion and Conclusion. In addition to its simplified implementation, the developedGBM strategy can be considered as a generative tool to provide the designer with severalsolutions. The shape optimization is considered as a numerical validation of the impor- tance of the different resulting RBTO layouts.