• CN：11-2187/TH
  • ISSN：0577-6686
机械工程学报

机械工程学报 ›› 2022, Vol. 58 ›› Issue (1): 179-189.doi: 10.3901/JME.2022.01.179

• 数字化设计与制造 • 上一篇    下一篇

扫码分享

基于产品特性关联和设计变更传播的向量空间重构模型

甘屹1,2, 何怡琪1, 高丽3,4, 李炳初1   

  1. 1. 上海理工大学机械工程学院 上海 200093;
    2. 中央大学理工学部 东京 112-0003 日本;
    3. 上海理工大学图书馆 上海 200093;
    4. 上海理工大学光电与计算机学院 上海 200093
  • 收稿日期:2021-01-22 修回日期:2021-08-25 出版日期:2022-01-05 发布日期:2022-03-19
  • 通讯作者: 甘屹(通信作者),男,1974年出生,博士,教授,博士研究生导师。主要研究方向为产品优化设计,智能制造技术,智能控制过程建模及优化算法。E-mail:ganyi@usst.edu.cn
  • 作者简介:何怡琪,女,1997年出生,硕士研究生。主要研究方向为产品优化设计,算法与过程建模。E-mail:193711563@st.usst.edu.cn
  • 基金资助:
    国家自然科学基金资助项目（51375314）。

Vector Space Reconstruction Model Based on Product Characteristic Linkage and Design Change Propagation

GAN Yi1,2, HE Yiqi1, GAO Li3,4, LI Bingchu1   

  1. 1. School of Mechanical Engineering, University of Shanghai for Science and Technology, Shanghai 200093;
    2. Faculty of Science and Engineering, Chuo University, Tokyo 112-0003, Japan;
    3. Library Department, University of Shanghai for Science and Technology, Shanghai 200093;
    4. School of Optoelectronics and Computer, University of Shanghai for Science and Technology, Shanghai 200093
  • Received:2021-01-22 Revised:2021-08-25 Online:2022-01-05 Published:2022-03-19

PDF

237

摘要： 一个有效且可控的产品特性设计变更可改善产品质量，促进产品创新，满足顾客需求。在产品设计过程中，由于产品特性间联系错综复杂，某个特性设计发生变化就有可能导致其他特性的一系列变化，从而引发变更传播及传播路径的重新规划。为此，基于向量空间模型（Vector space model，VSM）重构产品特性关联集（Product characteristic linkage set，PCLS）和设计变更工艺集（Design change technology set，DCTS），通过相似度计算获得PCLS-DCTS两集变更影响域，缩小了设计变更传播范围，构建设计变更传播的向量空间重构模型（Vector space reconstruction model，VSRM）；针对产品特性设计变更传播强度计算复杂度问题，利用拉格朗日松弛算法（Lagrangian relaxation algorithm，LRA）对VSRM变更影响域进行传播路径优化，并更新对偶函数下界以加快模型的解算速度；最后，以超声刀换能器的特性设计变更实验为例，实验结果表明，VSRM结合LRA方法能够快速有效的筛选出最优设计变更传播路径，可为产品设计变更传播机制和传播路径优化等相关工程领域提供决策指导。

关键词: 产品特性关联集, 设计变更工艺集, 设计变更传播, 拉格朗日松弛算法, 向量空间重构模型

Abstract: An effective and controllable product characteristic design change can improve product quality, promote product innovation, and meet customer needs. In the product design process, due to the intricate connections between product characteristics, a change in the design of one characteristic may lead to a series of changes in other characteristics, thereby triggering the propagation of changes and the re-planning of the propagation path. To this end, reconstructs the Product characteristic linkage set (PCLS) and Design change technology set (DCTS) based on the Vector space model (VSM). The two sets of change influence domains for PCLS-DCTS are obtained through similarity calculation, which reduces the scope of design change propagation, and constructs the Vector space reconstruction model (VSRM) of design change propagation; Aiming at the computational complexity of the propagation intensity of product characteristic design changes, the Lagrangian relaxation algorithm (LRA) is used to optimize the propagation path of the VSRM change influence domain. And update the lower bound of the dual function to speed up the solution of the model. Finally, take the characteristic design change experiment of the ultrasonic knife transducer as an example. Experimental results show that VSRM combined with LRA can quickly and effectively screen out the optimal design change propagation path. It can provide decision-making guidance for related engineering fields such as product design change propagation mechanism and propagation path optimization.

Key words: product characteristic linkage set, design change technology set, design change propagation, lagrangian relaxation algorithm, vector space reconstruction model

中图分类号: 

版权所有 © 2019 《机械工程学报》编辑部 | 京ICP备05055788号-30