Analysis And Optimamization For Anti-bending Behavior Of Hat-shaped Thin-walled Beams

As regulations about pedestrian safety and auto emission become more and more stringent, vehicle lightweight taking pedestrian protection into consideration has drawn much attention. As traditional absorption components, hat-shaped thin-walled beams generally achieve bending deformations in a car crash. Therefore, it is of important realistic significance to study the bending performance of these beams.Firstly, the anti-bending performances of single-hat thin-walled beams with square and trapezoidal cross section were numerically investigated using finite element models validated by bending test. The results show that specimens with trapezoidal cross section perform better in the bending resistant behavior than those with square section. To further improve bending behavior of trapezoidal section beams, the multi-objective optimization technique based on RBF response surface model and non-dominated sorting genetic algorithm Ⅱ (NSGA-Ⅱ) has been implemented. As a comparison, the same optimization processes were performed on specimens with square section. The studies indicate that both square and trapezoidal sections have better anti-bending behavior than original ones; meanwhile, Pareto front of the latter is closer to Utopian point.Secondly, a double-hat thin-walled beam of steel-aluminum hybrid materials (the aluminum upper hat and the steel lower hat, called as Al-steel Hat) is proposed to increase the energy absorption ability while reducing the initial peak force for the purpose of satisfying both the requirements for vehicle crashworthiness in high-speed crashes and the regulations protecting pedestrians and the structure itself from injuries during low speed collisions. The bending resistant behavior of Al-steel Hat is numerically explored using the validated finite element simulation methods and compared with tHat of counterparts with single material (called as Al Hat and Steel Hat) of identical geometry dimensions. The results show that Al-steel Hats perform best in the crashworthiness behavior under lateral impact. Parametric studies are conducted to investigate the influences of geometric parameters on bending performance of Al-steel Hat, Al Hat and Steel Hat. It is found that the thickness and height of the upper hat and the thickness of the lower hat affect significantly the bending responses of Al-steel Hat as well as Al Hat and steel Hat. In addition, using the radial basis function (RBF) metamodel and non-dominated sorting genetic algorithm (NSGA-II), multi-objective optimizations of Al-steel Hat, Al Hat and Steel Hat are carried out to achieve the maximum E (energy absorption) and minimum Fpeak (the initial peak force). The optimization results reveal that Al-steel Hat can yield better Pareto solutions than the counterparts of single material.