基于粒子群算法的轿车车身多学科优化设计  

Multidisciplinary Design Optimization of Car Body Based on Particle Swarm Algorithm

作　　者：刘钊[1,2] 李晗 朱平[1,2] Liu Zhao;Li Han;Zhu Ping(Shanghai Jiao Tong University,State Key Laboratory of Mechanical System and Vibration,Shanghai 200240;Shanghai Key Laboratory of Digital Manufacture for Thin-walled Structures,Shanghai 200240)

机构地区：[1]上海交通大学,机械系统与振动囯家重点实验室,上海200240 [2]上海市复杂薄板结构数字化制造重点实验室,上海200240

出　　处：《汽车工程》2018年第3期251-258,共8页Automotive Engineering

基　　金：国家自然科学基金(11772191)和国家自然科学基金青年科学基金(51705312)资助。

摘　　要：轿车车身结构设计过程中,设计变量多、非线性强、学科间耦合关系复杂,常用的多学科优化策略和优化算法无法有效寻得可行的优化解。本文中针对此问题,基于灵敏度分析方法与数据挖掘技术,采用粒子群优化算法;结合加强协同优化理论,形成了基于改进粒子群算法的轿车车身多学科优化设计体系,并将其运用到车身侧围结构轻量化设计中。结果表明,在满足各项结构性能指标的前提下,实现减质量最多达14.6%。In the process of car body structure design,due to large amount of variables,strong nonlinearity and the complex coupling relations between different disciplines,the common-used multidisciplinary design optimization(MDO)strategy and algorithm can not effectively get feasible optimization solutions.Aiming at this problem and based on sensibility analysis and data mining techniques in this paper,particle swarm optimization(PSO)algorithm is adopted and incorporated with the theory of enhanced collaboration optimization,a MDO system for car body based on modified PSO algorithm is formed and applied to the lightweight design of side panel structure.The results show that on the premise of meeting all structural performance indicators,a mass reduction of 14.6%at most is achieved.

关 键 词：车身 多学科优化设计 粒子群优化 数据挖掘 侧围 结构轻量化设计 

分 类 号：U463.82[机械工程—车辆工程]