Pro/E在动平衡分析自动优化过程中收敛性质的探讨  

Research on Convergence Properties of Pro/E in Automatic Optimization Process of Dynamic Balance

作　　者：黄建国[1] 程翔宇[2] 冯辉[2] HUANG Jianguo;CHENG Xiangyu;FENG Hui(Department of Basic Sciences,Anhui Water Conservancy Technical College,Hefei 231603,China;No.38 Research Institute of China Electronic Technology Group Corporation,Hefei 230088,China)

机构地区：[1]安徽水利水电职业技术学院基础部,安徽合肥231603 [2]中国电子科技集团公司第三十八研究所,安徽合肥230088

出　　处：《宿州学院学报》2020年第10期81-84,共4页Journal of Suzhou University

基　　金：国家国际科技合作专项(2011DFA12430);中国电子科技集团公司技术创新基金项目(JJ11304)。

摘　　要：多目标优化的路径选择过程因不能自动选择优化路径,而导致优化收敛过程缓慢,甚至无效的死循环运算难以避免。高速机构的精确动平衡问题,需要同时解决力和力矩的两个优化目标,这在基于Pro/E的自动优化运算过程中,基本无法得到理想解。为了解决这类通过计算机自动优化难以得到正确解的多目标优化问题,提出了一种把多目标优化的运算求解过程转化为单目标运算求解的方法:经过预先计算,规定好优化路径,再按照规定的优化路径,同时借助计算机遗传运算,进行自动优化求解。并从理论和实践过程中验证了该方法的收敛性,有效解决了同类多目标优化路径选择的NP-Hard问题。The path selection process of multi-objective optimization cannot automatically select the optimal path which leads to a slow optimization convergence process and even the invalid dead-cycle operation is hard to avoid.The problem of dynamic balancing of high speed spin component needs to solve minimizing force and minimizing moment simultaneously.In the process of automatic optimization operation based on Pro/E,the ideal solution cannot be obtained.Facing this difficult circumstance,we put forward a solution which converts multi-objective optimization operation into single-objective operation.After pre-calculation,the optimized path is specified.Then by use of gene-tic operation on the computer,the optimization result can be calculated automatically.We verify the effectiveness of the method in two ways of theory and practice.The NP-Hard problem of optimal path selection is solved effectively.

关 键 词：动平衡 多目标优化 太赫兹成像 

分 类 号：TH113.25[机械工程—机械设计及理论] TH113.24