• CN:11-2187/TH
  • ISSN:0577-6686

机械工程学报 ›› 2017, Vol. 53 ›› Issue (6): 166-175.doi: 10.3901/JME.2017.06.166

• 交叉与前沿 • 上一篇    下一篇

大功率远程射雾器结构参数多目标智能协同优化*

陈波1, 高殿荣1, 杨超1,2, 毋少峰1, 张光通1   

  1. 1. 燕山大学机械工程学院 秦皇岛 066004;
    2. 秦皇岛首创思泰意达环保科技有限公司 秦皇岛 066004
  • 出版日期:2017-03-20 发布日期:2017-03-20
  • 作者简介:陈波,男,1990年出生,博士研究生。主要研究方向为流体传动与控制。E-mail:flygo230@163.com高殿荣(通信作者),男,1962年出生,博士,教授,博士研究生导师。主要研究方向为流体传动与控制、复杂流场计算与可视化、新型流体元件等。E-mail:gaodr@ysu.edu.cn
  • 基金资助:
    * 国家自然科学基金(51375421),国家重点研发计划(2016YFC0206000)和河北省研究生创新(2016SJBS008)资助项目; 20160312收到初稿,20170115收到修改稿;

Multi-objective Intelligent Collaborative Optimization of Structure Parameters for High-power Remote Sprayer

CHEN Bo1, GAO Dianrong1, YANG Chao1,2, WU Shaofeng1, ZHANG Guangtong1   

  1. 1. College of Mechanical Engineering, Yanshan University, Qinhuangdao 066004;
    2. Qinhuangdao Capital Starlight Environmental Technology Co., Ltd., Qinhuangdao 066004
  • Online:2017-03-20 Published:2017-03-20

摘要:

为了探究主要结构参数对大功率远程射雾器的射雾距离和射雾效率的影响规律,得出最优设计方案,提出基于正交试验矩阵分析法、反向传播(Back propagation, BP)神经网络算法和遗传算法的多目标智能协同优化方法,利用计算流体动力学(Computational fluid dynamics, CFD)的方法建立正交试验数据库,采用多目标智能协同优化方法对关键结构参数(叶片与轮毂夹角θ、径向间隙h、导流叶片入角α、导流叶片弦长b、导流叶片个数n和射雾风筒倾斜角β)进行协同优化,并结合物理样机进行试验验证。结果表明,各结构参数对射雾距离和射雾效率的综合影响主次顺序依次为β>α>h>θ>b>n,最优结构参数为θ=47.8°、h=3.6、α=15.2°、b=261.6、n=7、β=1°,优化后的射雾距离提升了18.92%,射雾效率提升了12.94%,试验结果与数值计算优化结果基本吻合。该研究结果可为远程喷雾设备的设计和试验提供一定的参考。

关键词: BP神经网络, 射雾器, 遗传算法, 正交试验, 多目标优化

Abstract:

In order to explore the influence of main structure parameters on spray distance and spray efficiency of high-power remote sprayer, and find out the optimal design scheme, a multi-objective intelligent collaborative optimization method is proposed. Using numerical simulation method to establish orthogonal test database, the multi-objective intelligent collaborative optimization method which includes multi-objective orthogonal matrix method, BP (Back Propagation) neural network and genetic algorithms is applied to main structure parameters. The parameters include setting angle of bladeθ, the radial clearance of blade and spray barrelh, inlet angle of guide vaneα, chord length of guide vaneb, numbers of guide vanen, inclination angle of spray barrelβ. A prototype is produced according to the optimal results and a verification test is carried out. It is concluded that the primary and secondary sequence of structure parameters isβ>α>h>θ>b>n, which could reflect the comprehensive influence on spray distance and spray efficiency, the optimal structure parameters of sprayer areθ=47.8°, h=3.6, α=15.2°, b=261.6, n=7, β=1°, the spray distance is increased by 18.92% and the spray efficiency is increased by 12.94% through the multi-objective intelligent collaborative optimization method, the experimental test results are consist with the optimal result of numerical calculation. The research results provide reference for the design and experiment of remote sprayer device.

Key words: BP neural network, genetic algorithms, orthogonal test, sprayer, multi-objective optimization