锻压技术

基于神经元网络的薄壁筒滚珠旋压成形缺陷诊断

英文标题：Diagnosis of defects in ball spinning deformation of thin-walled tubular part based on ANN

作者：江树勇薛克敏李春峰张军

单位：哈尔滨工程大学工程训练中心合肥工业大学材料科学与工程学院哈尔滨工业大学材料科学与工程学院黑龙江哈尔滨哈尔滨工程大学工程训练中心黑龙江哈尔滨150001 安徽合肥230009 150001 黑龙江哈尔滨150001

关键词：滚珠旋压强力旋压神经元网络铝合金

分类号：TG376

出版年,卷(期):页码：2006,31(3):79-83

摘要：

作为一种连续局部塑性成形工艺,滚珠旋压被应用于制造高强度、高精度的纵向内筋薄壁筒形件。通过使用铝合金作为旋压材料,在实验的基础上分析了滚珠旋压过程中金属材料非稳定流动的基本原理及旋压件产生表面质量缺陷的原因。以人工神经元网络为基础,对旋压件的表面质量缺陷进行了预测。实验证明,神经元网络能够精确地诊断旋压件的表面质量缺陷。

As a successively and locally plastic deformation process,ball spinning is applied in order to manufacture high-strength and high-precision thin-walled tubular part with longitudinal inner ribs.By using aluminum alloy as spinning material,based on the experiments,not only the basic principle with respect to non-steady flow of metal material in ball spinning,but also the reasons for surface quality defects of the spun parts are analyzed.On the basis of artificial neural networks(ANN),the surface quality defects of the spun parts is predicted.Experiments have proved that ANN can predict and diagnose the surface quality defects of the spun part successfully.

基金项目：

参考文献：

[1]Wong C C,Dean T A,Lin J.A review of spinning,shearforming and flow forming process[J].International Journalof Machine Tools and Manufacture,2003,43(14):1419 1435.
[2]Rotarescu M I.A theoretical analysis of tube spinning usingballs[J].Journal of Materials Processing Technology,1995,54(1 4):224 229.
[3]李茂盛,康达昌,张士宏,等.滚珠旋压工艺中成形区接触压力的分析计算[J].材料科学与工艺,2004,12(2):125 128
[4]Basheer I A,Hajmeer M.Artificial neural networks:funda-mentals,computing,design,and application[J].Journal ofMicrobiological Methods,2000,43(1):3 31.
[5]Quinlan P T.Structural change and development in real andartificial neural networks[J].Neural Networks.1998,11(4):577 599.
[6]JIANG Shu-yong,LI Ping,XUE Ke-min.Application ofBPANN in spinning deformation of thin-walled tubular partswith longitudinal inner ribs[J].Journal of Central South U-niversity of Technology,2004,11(1):27 30.
[7]Kim D J,Kim B M.Application of neural network and FEMfor metal forming processes[J].International Journal of Ma-chine Tools and Manufacture,2000,40(6):911 925.
[8]Gunasekera J S,Jia Z J,Malas J C,et al.Development of aneural network model for a cold rolling process[J].Engi-neering Application of Artificial Intelligence.1998,11:597603.

服务与反馈：

【文章下载】【加入收藏】