北京航空航天大学学报 ›› 2015, Vol. 41 ›› Issue (9): 1600-1607.doi: 10.13700/j.bh.1001-5965.2015.0002

• 论文 • 上一篇    下一篇

参数化产品族递进式优化设计方法

魏巍1, 冯毅雄2, 程锦2   

  1. 1. 北京航空航天大学 机械工程及自动化学院, 北京 100191;
    2. 浙江大学 流体动力与机电系统国家重点实验室, 杭州 310027
  • 收稿日期:2015-01-04 出版日期:2015-09-20 发布日期:2015-09-29
  • 通讯作者: 魏巍(1982—),男,辽宁沈阳人,讲师,weiwei@buaa.edu.cn,主要研究方向为产品平台设计、产品族设计、制造业信息化. E-mail:weiwei@buaa.edu.cn
  • 基金资助:
    国家自然科学基金(51205010,51322506); 中央高校基本科研业务费专项资金; 浙江省自然科学基金(LR14E050003)

Parametric product family progressive optimization design approach

WEI Wei1, FENG Yixiong2, CHENG Jin2   

  1. 1. School of Mechanical Engineering and Automation, Beijing University of Aeronautics and Astronautics, Beijing 100191, China;
    2. The State Key Laboratory of Fluid Power Transmission and Control, Zhejiang University, Hangzhou 310027, China
  • Received:2015-01-04 Online:2015-09-20 Published:2015-09-29

摘要: 基于产品平台技术,提出了基于混合进化算法的参数化产品族递进式优化设计方法.建立了参数化产品族的递进式优化设计流程,提出改进的强度Pareto进化算法(SPEA2+)与非支配排序遗传算法(NSGA-II)相结合的多目标混合进化算法对参数化产品族设计问题进行优化,混合进化算法使用了两类种群进行求解,解决了同步进化带来的数据扰动问题.在参数化产品族递进式优化设计过程中,首先优化产品族设计平台,建立参数化产品族设计问题的多目标优化数学模型,通过产品设计参数的敏感度分析和变差指数计算,划分产品平台的设计常量和设计变量,形成稳健的产品平台来获得最优参数.然后对产品族中实例产品的多个性能进行优化,在已有的产品平台基础上,优化设计变量的取值.最后,以电动机产品族的递进式优化设计过程为例证明了该方法的有效性和适用性.

关键词: 产品族设计, 产品平台, 递进式设计, 混合进化算法, 敏感度, 变差指数

Abstract: Parametric product family progressive optimization design approach is proposed based on product platform technique. The progressive optimization design process was constructed, the optimization of product family was carried out progressively using multi-objective mix-evolution algorithm. The mix-evolution algorithm used two kinds of populations to avoid the data perturbation problem. The strength Pareto evolutionary algorithm 2+and non-dominated sorting genetic algorithm-II were used progressively in the product platform design and product instance optimization. In the product family progressive optimization, the product platform was identified firstly. The multi-objective optimization mathematics optimization model of parametric product family was constructed, the sensibility of design parameter was analyzed and the diversity factor was calculated. The product platform constants and variables were divided. As a result, the product platform was constructed. Then the performances of each individual instance product were optimized in the robust product platform to get the optimal design parameter. Finally, the design of electromotor product family was used as an example to certify the proposed method's effectiveness and applicability.

Key words: product family design, product platform, progressive design, mix-evolution algorithm, sensibility, diversity factor

中图分类号: 


版权所有 © 《北京航空航天大学学报》编辑部
通讯地址:北京市海淀区学院路37号 北京航空航天大学学报编辑部 邮编:100191 E-mail:jbuaa@buaa.edu.cn
本系统由北京玛格泰克科技发展有限公司设计开发