Application of a weighted graph representation method to the crashworthiness optimization of subway collision frame structures

Abstract

The collision frame structure is the most important subway structure for energy absorption and impact load transfer during foreign object collision. To improve the crashworthiness of the collision frame structure, a parameter optimization method based on a weighted graph representation method was proposed. Based on weighted graph theory, an equivalent mathematical model of the collision frame structure was established. Combined with the Belytschko beam element method, which is a large deformation calculation technique, a parametric finite element model (FEM) considering plastic mechanical properties was constructed, which was validated by a detailed finite element model and a quasi-static crush test. Based on the practical requirements, the crush displacement, the intrusion space and the mean crush force (MCF) were determined as constraints. The structural total mass was considered as the objective. The optimized structural parameters and key performance indicators were obtained and compared by the genetic algorithm (GA), Colliding bodies optimization (CBO), Enhanced colliding bodies optimization (ECBO). The results showed that the GA algorithm had the best optimization objective. The comparison results showed that the optimized collision frame structure could meet the crashworthiness requirements, while the structural mass was reduced from the initial 168.92 kg to 97.33 kg, thus giving a reduction rate of 42.38 %.