Crash performance and multi-objective optimization of a gradual energy-absorbing structure for subway vehicles
ping xu,chengxing yang,yong peng,shuguang yao,dehong zhang,benhuai li
DOI: https://doi.org/10.1016/j.ijmecsci.2016.01.001
IF: 7.3
2016-01-01
International Journal of Mechanical Sciences
Abstract:To improve the crash performance of subway vehicles, this study presents an investigation of a newly designed gradual energy-absorbing structure subjected to impact loads using a test trolley for experimental and numerical simulations, as referenced in EN15227:2008. The structure is composed of a front-end beam, rear-end plate, outer-side beams, inner-side beams, cross beams and longitudinal beams. In this work, a finite element model (FEM) was established and effectively verified using experimental data. To evaluate the effect of the thicknesses of the longitudinal beams (t1), outer-side beams (t2) and inner-side beams (t3) on the energy absorption (EA) and peak crushing force (Fp), sensitivity study was performed using extreme difference (R) analysis. Then, a polynomial response surface method (PRSM) was employed to formulate the output responses of EA and Fp with the input parameters of t1, t2 and t3. In order to optimize the crash performance of the structure, a Non-dominated Sorting Genetic Algorithm (NSGA) was used in a multi-objective optimization to achieve the maximum EA and minimum Fp values. The optimal EA and Fp with the associated thickness parameters (t1, t2, and t3) of the gradual energy-absorbing structure were obtained in the form of Pareto fronts. Additionally, the Pareto solutions derived from the surrogate models of the quadratic, cubic and quartic polynomials eventually converge at intersection D (i.e., Fp=7.2940×103kN, EA=1.3511×103kJ).