期刊首页 优先出版 当期阅读 过刊浏览 作者中心 关于期刊 English

《中国工程科学》 >> 2008年 第10卷 第5期

基于人工神经网络的多处损伤加筋板剩余强度预测

1. 海军航空工程学院研究生队,山东烟台264001;

2. 海军航空工程学院青岛分院,山东青岛266041

资助项目 :国家自然科学基金资助项目(50675221);总装“十一五”预研课题(513270301) 收稿日期: 2006-12-19 发布日期: 2008-05-15 16:44:41.000

下一篇 上一篇

摘要

用BP神经网络算法对多处损伤加筋板的剩余强度数据进行训练学习,将预测值和3种经典分析方法的计算值与实验值进行对比,结果表明,ANN法预测值与实验值吻合得最好,LMC修正法和WSU3修正法次之,Swift塑性区连通法最差。最后用所建立的BP网络对不同主裂纹半长和韧带长度的剩余强度进行了预测,结果发现,在其他参数不变的情况下,不管是双筋条还是三筋条加筋板,剩余强度总是随主裂纹半长的增加而成线性降低,随韧带长度的增加而成线性增加,但双筋条加筋板比三筋条加筋板对主裂纹半长和韧带长度的变化更加敏感。

图片

图1

图2

图3

图4

图5

图6

图7

图8

图9

参考文献

[ 1 ] National Transportation Safety Board.Aircraft Accident Report [ R] .Washington D.C.:Report NTSB /AAR -89 /03 ,1989

[ 2 ] Swift T.Widespread fatigue damage monitoring issues and con- cerns [ A] .5th International Conference on Structure Airworthi- ness of New and Aging Aircrat [ C] .Tokyo,1993 链接1

[ 3 ] Smith B L, Hijazi A L, Haque A K M, et al.Modified linkup models for determining the strength of stiffened panels with multi- ple site damage [ A] .Proceedings of the FAA /NASA Symposium on the Continued Airworthi - ness of Aircraft Structures [ C ] . Houston, 1999 ,555 -566

[ 4 ] Smith B L, Saville P A, Mouak A, et al.Strength of 2024 -T3 a- luminum panels with multiple site damage [ J] .Journal of Air- craft, 2000 , 37 ( 2 ) : 325 -331 链接1

[ 5 ] Alal Hijaxi.Residual Strength of Thin -sheet Aluminum Panels with Multiple Site Damage [ D].Chita: PHD, Wichita State University, 1999

[ 6 ] Ingram J, Kwon Y, Duffe K, et al.Residua strength analysis of skin splices with multiple site damage [ A ] . The 2nd Joint NASA /FAA /-DOD Conference on Aging Aircraft [ C ] .San Francisco, 1998 链接1

[ 7 ] Thomson D, Hoadly D, Mchatton J.Load Tests of Flat and Curved Panels with Multiple Cracks [R].Foster Miller Inc,1993

[ 8 ] DeWit R, Fields R, Mordfin L, et al.Fracture behavior of large -scale thin -sheet aluminum alloy [ A] .National Institute of Standards and Technology[ C] .Los Angeles,1995

[ 9 ] Smith B, Mouak A, Savill P, et al.Improved engineering methods for determining the critical ctrength of aluminum panels with mul- tiple site damage [ A] .The 2nd Joint NASA /FAA /DOD Confer- ence on Aging Aircraft [ C] .San Francisco,1988

[10] 袁曾任.人工神经网络及其应用[M].北京:清华大学出版社,1999

相关研究