基于同一车体结构模型的不同优化算法对比研究  

Comparative Study of Different Optimization Algorithms Based on the Same Car Body Structure Model

作　　者：石晓飞 李斌 邵振伟 刘肖鹏 王震宇 SHI Xiaofei;LI Bin;SHAO Zhenwei;LIU Xiaopeng;WANG Zhenyu(Guangzhou Railway(Group)Company,Guangzhou EMU Depot,Guangzhou Guangdong 511483,China)

机构地区：[1]中国铁路广州局集团有限公司广州动车段,广东广州511483

出　　处：《铁道运营技术》2024年第3期4-8,共5页Railway Operation Technology

摘　　要：高速列车工程优化设计中有多种优化算法,解决实际优化问题时选择合适的算法能提高优化设计效率。以某一高速列车铝合金车体为模型,通过有限元分析计算整备状态下的车体一阶垂弯频率较低。在灵敏度分析的基础上,应用正交试验对各个变量进行主效应分析,筛选出主要变量作为优化模型的设计变量。以提高一阶垂弯频率为目的,基于近似模型分别采用可行方向法和遗传算法进行优化求解,使用可行方向法优化后的车体一阶垂弯振动频率提升2.4%,二阶菱形振动提升1.9%,车体钢结构质量增重0.15%;使用遗传算法优化后车体一阶垂弯振动频率提高了4.2%,车体的二阶菱形振动频率提高2.5%,车体钢结构质量增重1.66%。对优化结果进行对比分析,两种不同算法的优化结果侧重点有所不同,可根据不同实际需求选择相应的优化算法,从而达到优化目的。There are various optimization algorithms in the optimization design of high-speed train engineering.When solving actual optimization problems,choosing the right algorithm can improve the optimization design ef⁃ficiency.Based on the model of an aluminum alloy body of a high-speed train,the first order sag frequen⁃cy of the body is calculated by finite element analysis.On the basis of sensitivity analysis,orthogonal test is applied to conduct principal effect analysis on each variable,and the main variables were selected as the design variables for optimizing the model.In order to improve the first-order bending frequency,the feasible direction method and genetic algorithm were used to optimize the solution based on the approxi⁃mate model,respectively.The first-order bending vibration frequency and second-order rhomboid vibra⁃tion are improved by 2.4%and 1.9%respectively,and the weight of the steel structure of the train body was increased by 0.15%.After using genetic algorithm optimization,the first-order bending vibration fre⁃quency and second-order rhomboid vibration are improved by 4.2%and 2.5%,and the weight of the steel structure of the train body was increased by 1.66%.The comparison and analysis of the optimization results show that the two different algorithms have different emphasis on the optimization results,and the corresponding optimization algorithm can be selected according to different actual needs,so as to achieve the optimization purpose.

关 键 词：有限元分析 近似模型 可行方向法 遗传算法 对比分析 

分 类 号：U271.91[机械工程—车辆工程]