Sampling strategy using genetic algorithms in the optimization of a technical design

摘要

A sample determination strategy using genetic algorithms (GA) in engineering design optimization is described. A product must be designed and optimized with a set of design variables, goals and constraints. An appropriate number of sample experimental design (DOE) samples is then identified so that each point represents a specific or unique combination of design variables. The sample selection strategy is based on genetic algorithms. A computer aided design analysis (CAE) or analysis (eg, finite element analysis, finite difference analysis, mesh-free analysis, etc.) is performed for each of the samples during the GA-based sample selection process. A meta-model is generated to approximate the results of the CAE analysis on all DOE samples. If the meta-model is satisfactory (eg, its accuracy is within a tolerance), an optimized "best" design can be found using the metamodel as a function evaluation device for the optimization process. Finally, a CAE analysis is performed to verify the optimized "best" design.