Multiobjective and multicollision scenario reliability-based design optimization of honeycomb-filled composite energy-absorbing structures for subways

A honeycomb-filled composite energy-absorbing structure (HCES) installed at the front end of a subway has an extensive range of application prospects in various crashworthy buffer structures due to its excellent energy-absorbing performance. Research to date on the crashworthiness optimization of energy-absorbing structure has mainly focused on a single collision scenario. However, in train collision accidents, the collision scenarios are diverse and unpredictable, implying that the optimal design under a certain collision scenario may no longer be applicable under other collision scenarios. Additionally, in actual manufacturing, there may be uncertainties in the design parameters. Unfortunately, uncertain structural optimization problems solved with deterministic optimization methods may lead to structural instability or even failure. To this end, this study formulates a multiobjective and multicollision scenario reliability-based design optimization that combines a radial basis function, Monte Carlo simulation (MCS), NSGA-II and the order preference by similarity to an ideal solution (TOPSIS) to seek an optimal reliability design for HCESs that meets the requirements of multiple collision scenarios in European standard EN 15227. The optimization results show that the presented reliability-based optimization method has fair effectiveness and can enhance the robustness of an HCES under multiple collision scenarios, implying that the proposed approach can provide meaningful guidelines for the structural crashworthiness design of subways.