Crashworthiness design of a steel-aluminum hybrid rail using multi-response objective-oriented sequential optimization

Hybrid structures with different materials have aroused increasing interest for their lightweight potential and excellent performances. This study explored the optimization design of steel-aluminum hybrid structures for the highly nonlinear impact scenario. A metamodel based multi-response objective-oriented sequential optimization was adopted, where Kriging models were updated with sequential training points. It was indicated that the sequential sampling strategy was able to obtain a much higher local accuracy in the neighborhood of the optimum and thus to yield a better optimum, although it did lead to a worse global accuracy over the entire design space. Furthermore, it was observed that the steel-aluminum hybrid structure was capable of decreasing the peak force and simultaneously enhancing the energy absorption, compared to the conventional mono-material structure.