欢迎访问《图学学报》 分享到:

图学学报

• 数字化设计 • 上一篇    下一篇

基于协同进化算法的分支线缆自动布局设计方法

  

  • 出版日期:2016-02-26 发布日期:2016-02-26

Branch Cable Automatic Routing Based on Co-evolutionary Algorithm

  • Online:2016-02-26 Published:2016-02-26

摘要: 针对分支线缆布局设计中分支点难以确定的问题,提出基于协同进化算法的分支
线缆自动布局设计方法。首先建立分支线缆的线束模型,基于线缆的线束模型信息将分支线缆
分解为多个一对一的单根线缆,每个单独的线缆代表一个种群;在此基础上采用改进的快速扩
展随机树算法求解单根线缆的路径,然后基于协同进化的思想对分支线缆的分支点进行寻优,
通过种群间相互影响适应度的评价使得分支线缆的布局结果达到最优;最后对最终优化得到的
路径点进行拟合,从而获得线缆作为布局设计结果并输出。设计并开发了线缆自动布局设计软
件原型系统,进行算例测试与实例应用,验证方法的可行性。

关键词: 分支线缆, 自动布局设计, 协同进化, 快速扩展随机树

Abstract: Facing the problem that the middle forked point is difficult to determine which exists in the
branch cable routing, a branch cable automatic routing method is proposed based on co-evolutionary
algorithm. Firstly, the branch cable harness model is established. The approach divides the whole
system of branch cable into a few single cables based on the cable harness model, and every single
cable represents a population. Improved RRT (rapidly-exploring random trees, RRT) algorithm is
used to obtain initial path of every single cable. Then middle forked points is optimized based on the
idea of co-evolution, which is affected by other single cables during its evolution, and the branch
cable routing optimized obtained. Finally, the final optimization of path points obtained were fitted to
obtain the cable as a result. A 3D automatic routing prototype system is developed and some
experiments are applied to verify the efficiency of the method.

Key words: branch cable, automatic routing, co-evolutionary algorithm, rapidly-exploring random
tree algorithm