Design optimization of a novel NPR crash box based on multi-objective genetic algorithm

Possessing the unique properties of lower mass and higher performances, the structure with Negative Poisson’s Ratio (NPR) can be widely used in aerospace and vehicle industry. By combing the NPR structure filled core and the traditional crash box, a novel NPR crash box is first proposed in this work to improve the performances of the crash box. The performances of the novel NPR crash box are fully studied by comparing to the traditional crash box and the aluminum foam filled crash box. A parameterized model of the NPR crash box, which integrates the design parameters of the basic NPR cell structure, is built to improve the analysis and optimization efficiency, the accuracy of the parameterized model is also verified by comparing to traditional FEM model. Multi-objective optimization model of the NPR crash box is established by combining the parameterized model, optimal Latin square design method and response surface model approach. Non-dominated sorting genetic algorithm-II (NSGA-II) is then applied to optimize the design parameters of the basic NPR cell structure to improve the performances of the NPR crash box. The results indicate that the novel NPR crash box can improve the performances of the crash box remarkably and the combination of parameterized model and multi-objective genetic algorithms optimize the NPR crash box efficiently. The presented new method also serves as a good example for other application and optimization of NPR structure.