• CN:11-2187/TH
  • ISSN:0577-6686

机械工程学报 ›› 2018, Vol. 54 ›› Issue (19): 197-203.doi: 10.3901/JME.2018.19.197

• 数字化设计与制造 • 上一篇    下一篇

基于MOPSO的航空发动机分支管路多目标布局优化

柳强, 毛莉   

  1. 辽宁石油化工大学信息与控制工程学院 抚顺 113001
  • 收稿日期:2017-11-09 修回日期:2018-04-28 出版日期:2018-10-05 发布日期:2018-10-05
  • 通讯作者: 柳强(通信作者),男,1982年出生,博士,副教授,硕士研究生导师。主要研究方向为复杂装备管路敷设优化及振动分析等。E-mail:neuliuqiang@163.com
  • 作者简介:毛莉,女,1992年出生,硕士研究生。主要研究方向为管路布局以及多目标优化。E-mail:993880336@qq.com
  • 基金资助:
    国家自然科学基金(51305192)、辽宁省高等学校杰出青年学者成长计划(LJQ2014037)和辽宁省自然科学基金(20170540589)资助项目。

Multi-objective Layout Optimization for Branch Pipe of Aero-engine Based on MOPSO

LIU Qiang, MAO Li   

  1. School of Information and Control Engineering, Liaoning Shihua University, Fushun 113001
  • Received:2017-11-09 Revised:2018-04-28 Online:2018-10-05 Published:2018-10-05

摘要: 分支管路的布局优化属于NP难问题,其多目标优化情况则更加复杂。针对航空发动机分支管路多目标敷设问题,以分支管路长度最小化、分支点数量最小化以及管路平滑度最优为优化目标,建立了基于避障Steiner树的分支管路多目标布局模型。考虑到模型的复杂性,设计基于多目标粒子群优化(Multi-objective particle swarm optimization,MOPSO)的模型求解算法。其中,以分支点数量和坐标作为决策变量;针对分支管路拓扑结构特点,提出一种分支管路平滑度计算方法,结合非支配排序和网格密度计算完成个体多目标评价;通过可视图和测地线处理约束条件;通过多目标粒子群进化计算求得Pareto解集。所建立的分支管路多目标布局模型及求解算法考虑了多端点情况、多目标优化以及避障约束。最后通过管路敷设算例验证了可行性。

关键词: 多目标粒子群优化, 分支管路, 管路敷设, 航空发动机

Abstract: The layout optimization of branch pipe is a NP hard problem, and the multi-objective case of which is more complex. The branch pipe routing problem is studied in the context of an aero-engine development, and a model of multi-objective routing for branch pipes is constructed based on Steiner tree, where pipe length, the number of branch points and pipeline smoothness are taken into consideration as optimal objectives. Considering the complexity of the problem, a multi-objective particle swarm optimization (MOPSO)-based layout method is designed to solve the constructed routing model. More specifically, the number and coordinates of branch points are selected as decision variables; according to the characteristic topological structure of branch pipes, a computation model of smoothness is presented, and the multi-objective evaluation is then completed by non-dominated sorting and grid density calculation; in addition, the visibility graph method and geodesic are integrated to handle routing constraints; The Pareto solution set is obtained through particle evolution. The presented routing model and method consider the cases of multiple pipe terminals, multi-objectives and obstacle-avoidance constraints. Finally, the feasibility of the proposed method is demonstrated by several numerical computations of branch pipe routing examples.

Key words: aero-engine, branch pipe, multi-objective particle swarm optimization, pipe layout

中图分类号: