标题：A hybrid multi-objective optimization approach for energy-absorbing structures in train collisions
作者：Zhang, Honghao; Peng, Yong; Hou, Lin; Tian, Guangdong; Li, Zhiwu
作者机构：[Zhang, Honghao; Peng, Yong; Hou, Lin] Cent S Univ, Sch Traff & Transportat Engn, Key Lab Traff Safety Track, Minist Educ, Changsha 410000, Hunan, Peo 更多
通讯作者：Peng, Yong;Tian, GD;Peng, Y
通讯作者地址：[Tian, GD]Shandong Univ, Sch Mech Engn, Jinan 250061, Shandong, Peoples R China;[Peng, Y]Minist Educ, Key Lab Traff Safety Track, Sch Traff & Transpor 更多
关键词：Multi-objective optimization; Energy-absorbing structure; Artificial bee; colony; Grey relational analysis; VIKOR; Train collision
摘要：Energy-absorbing structure, which is the most effective and direct protection component, is installed at the front of the head car. However, structural optimization problems of this structure still exist, e.g., multiple conflicting objectives and the non-uniqueness problem of optimization solutions. This study formulates a novel optimization framework combining the theory of multi-objective optimization and multi-criteria decision making and proposes a hybrid optimization approach (M-BGV) that combines multi-objective artificial bee colony (MOABC), best worst (BW) method, grey relational analysis (GRA) and visekriteri-jumsko kompromisno rangiranje (VIKOR), to solve the structural optimization problem for energy-absorbing structures in train collisions. MOABC is applied to determine the points that represent the optimal solutions, i.e., the Pareto set. The preferences for the conflicting objectives for the certain practice condition/structure can be calculated by the BW method. An integrated multi-criteria decision making approach that combines GRA and VIKOR is proposed to obtain the optimal solution via evaluating the solutions from the Pareto set. Subsequently, an empirical application of a multi-cell thin-walled aluminum energy-absorbing structure is applied to demonstrate that this utilized integrated methodology is valid and practical. The results prove that this approach provides an accurate and effective tool for the structural multi-objective optimization problem. (C) 2019 Elsevier Inc. All rights reserved.