China Mechanical Engineering

Previous Articles     Next Articles

Evolutionary Design of Mechanism Kinematics Chain of Genetic Programming

HUA Ertian1,2;ZHOU Meng1;LI Shenghui1;LU Hao1   

  1. 1.College of Mechanical Engineering, Zhejiang University of Technology, Hangzhou, 310023
    2.Zhejiang University of Water Resources and Electric, Hangzhou, 310018
  • Online:2019-08-25 Published:2019-08-27

遗传编程的机构运动链进化设计

华尔天1,2;周猛1;李生辉1;鹿浩1   

  1. 1.浙江工业大学机械工程学院,杭州,310023
    2.浙江水利水电学院,杭州,310018
  • 基金资助:
    国家自然科学基金资助项目(61722469)

Abstract: A new mechanism and the original mechanism of regenerative kinematics chain method had the same number of components and kinematic pairs.Genetic programming was proposed to replace the number synthesis, the generalization and embodiment processes of the regenerative motion chain method were retained, the essential components of original mechanism were analyzed and deter-mined, which ensured that the new mechanism satisfied the functional properties of the original mechanism, realized the evolutionary design of different number of components and kinematic pairs of the new mechanism and the original mechanism.The fitness evaluation function was designed based on the contents of information axiom of axiomatic design.The total number of components and the number of multiple auxiliary bars were used as the fitness measure factors, and a feasible solution with the smallest amount of information was obtained.The experimental results show the effectiveness of evolutionary design method of the mechanism kinematic chain based on genetic programming.

Key words: regenerative motion chain method, genetic programming, evolutionary design, fitness evaluation

摘要: 针对再生运动链法存在的新机构与原机构具有相同构件数和运动副数的局限,提出采用遗传编程替代数综合,保留了再生运动链法的一般化和具体化过程,分析确定原机构的必备构件,保证新机构满足原机构的功能特性,实现新机构与原机构不同构件数和运动副数的进化设计;基于公理设计中信息公理的内容设计适应度评价函数,以构件总数和多副杆数作为适应度衡量因子,得到具有最小信息量的可行解。通过实例验证,证明了基于遗传编程的机构运动链进化设计方法的有效性。

关键词: 再生运动链法, 遗传编程, 进化设计, 适应度评价

CLC Number: