欢迎访问《汽车安全与节能学报》，
汽车安全与节能学报

JASE ›› 2017, Vol. 08 ›› Issue (01): 30-37.DOI: 10.3969/j.issn.1674-8484.2017.01.003

• 汽车安全 • 上一篇    下一篇

改进粒子群优化算法在轿车车身轻量化设计中的应用

刘 钊1,2，朱平1,2*，李泽阳1,2   

  1. 1. 上海交通大学 机械系统与振动国家重点实验室，上海 200240，中国；
    2. 上海市复杂薄板结构数字化制造重点实验室，上海 200240，中国
  • 收稿日期:2016-11-29 出版日期:2017-03-23 发布日期:2017-03-23
  • 通讯作者: 朱平(1966—)，男( 汉族)，浙江，教授，E-mail: pzhu@sjtu.edu.cn。
  • 作者简介:第一作者 / First author ： 刘钊(1988—)，男( 汉族)，山东，博士，E-mail: kakalz@163.com。 通讯作者 / Corresponding author ： 朱平(1966—)，男( 汉族)，浙江，教授，E-mail: pzhu@sjtu.edu.cn。
  • 基金资助:

    国家自然科学基金项目(11372181)

Modified particle swarm optimization algorithm using in lightweight design of auto-body

LIU Zhao1,2, ZHU Ping *1,2, LI Zeyang1,2   

  1. 1. State Key Laboratory of Mechanical System and Vibration, Shanghai Jiao Tong University, Shanghai 200240, China;
    2. Shanghai Key Laboratory of Digital Manufacture for Thin-walled Structure, Shanghai 200240, China
  • Received:2016-11-29 Online:2017-03-23 Published:2017-03-23

PDF

1075

摘要：

为了提高粒子群优化算法（PSO）寻优过程中粒子种群之间的多样性，提出了一种基于停滞判断准则与粒子速度更新的改进PSO。在寻优过程中，当粒子群优化陷入停滞时，新的速度更新公式将激活，提高粒子种群之间的多样性。通过数学标准测试函数，进行了数学实验，对比改进PSO 与标准PSO 的优化能力。结合近似建模技术与合理的约束处理方法，将该算法运用到考虑100% 正面碰撞工况、40% 偏置碰撞工况、侧面碰撞工况、追尾碰撞工况以及顶压溃工况的车身轻量化设计中。结果表明：在全局寻优能力方面，运用改进PSO 优化得到的问题解均优于标准PSO。在满足各项结构性能指标的前提下，可减轻质量23.41 kg。这为轿车车身轻量化设计提供了可借鉴的方法。

关键词: 车身设计, 轻量化, 多种碰撞工况, 粒子群优化算法（PSO）, 停滞判断准则, 近似建模

Abstract:

A modified Particle Swarm Optimization (PSO) algorithm was built based on stagnation judging criterion and particle velocity updating equation to improve the diversity of particles during optimization procedure. A new velocity updating equations was activated to improve the diversity between particles while the particles optimization procedure fell into stagnation. The modified PSO algorithm was compared with the standard version of PSO based on a set of benchmark functions by mathematical experiments. Auto body lightweight optimization design was made by using the modified PSO incorporated with metamodeling technique and reasonable constrained handling method, considering 100% frontal impact, 40% frontal offset impact, side impact, rear impact and roof crash conditions, and under the premise of satisfying each structural performance index. The results show that the optimization solutions by modified PSO algorithm are superior to those by standard PSO with areduced mass of 23.41 kg. This will provide a referential method for light mass design of auto-body.

Key words: design of auto-body, lightweight, multiple impact conditions, particle swarm optimization (PSO), stagnation judging criterion , metamodeling