用于结构损伤识别的神经网络输入选取规则探究

英文篇名：Exploration of selection rules for inputs of neural network used on structural damage identification
中文刊名：四川建筑科学研究
英文刊名：Sichuan Building Science
作者：张育智 ; 何伟 ; 李乔 ; 单德山
英文作者：ZHANG Yuzhi1 ; HE Wei2 ; LI Qiao1 ; SHAN Deshan1(1.School of Civil Engineering ; Southwest Jiaotong University ; Chengdu 610031 ; China ; 2.College of Sichuan Architectural Technology ; Deyang 618000 ; China)
中文关键词：损伤识别 ; 神经网络 ; 网络输入 ; 模式识别
英文关键词：damage identification ; neural network ; network input ; mode identification
出版日期：2010-02-25
机构：西南交通大学土木工程学院;四川建筑职业技术学院;
年：2010
期：01
出版单位：四川建筑科学研究

摘要

将神经网络作为模式识别工具用于结构损伤位置识别时,其识别效果除了要受到网络隐层数目、各隐层神经元数目、神经元传递函数的形式、训练样本的数量与质量及训练方法的影响外,还会受网络输入性能的影响。在其他因素均相同的条件下,网络输入对网络性能起着决定性作用。为解决网络输入的选取问题,从网络功能、类别可分性和噪声的影响三个方面对网络输入的选取进行了分析研究,提出了用于结构损伤识别的神经网络输入选取的一般性规则,对采用神经网络处理模式识别问题具有参考价值。
When the diagnosis of damage location is taken as a problem of mode identification,neural network can be used as a pattern classifier.The identification effects will be affected by the number of the connotative layers and the nerve cells,the transfer function mode,the training samples,the training method and the performance of the inputs.Given the same conditions,the performance of the inputs is crucial to the performance of the neural network.To solve the problem of inputs sealection,three aspects:the function of the neural network,the separability of the inputs and the effect of noise,are taken into consideration.The selection rules of neural network inputs that have been put forward in this article is instructive to those who will use neural network to solve the problem of mode identification.

引文

[1]Worden K.Structural fault detection using a novelty measure,Jour-nalof Sound and Vibration,1997,201:85-101.
    [2]Ni YQ,Li H,Wang J Y,at el.Implementation issues of novelty de-tection technique for cable-supported bridges instrumented with along-term monitoring system,Proceedings of SPIE,2002,4704:225-236.
    [3]高赞明,孙宗光,倪一清.基于振动方法的汲水门大桥损伤检测研究[J].地震工程与工程振动,2001,21(4):117-124.
    [4]Ko J M,Ni Y Q,Chan,et al.Feasibility of damage detection of Ts-ing Ma Bridge using vibration measurements,Proceedings of SPIE,2000,3995:370-381.
    [5]Sohn H,Worden K,Farrar C R,et al.Novelty detection underchanging environmental conditions,Proceedings of SPIE,2001,4330:108-118.
    [6]Zhao J,DeWolf,John T.Sensitivity study for vibrational parametersused in damage detection.Journal of structural engineering,1999,125(4):410-416.
    [7]Cawley P,Adams R D.Location of defects in structures from meas-urements of natural frequencies,Journal of Strain Analysis,1979,14(2):49-57.
    [8]Hearn G,Testa R B.Modal Analysis for Damage Detection inStructures.Journal of Structural Engineering,1991,117(10):3042-3063.
    [9]陈素文.高层建筑人工神经网络损伤识别及安全性综合评判[D].上海:同济大学,2001.
    [10]Lam K F,Ko M J,Wong C W.Localization of damaged structuralconnections based on experimental modal and sensitivity analysis.Journal of Sound and Vibration,1998,210(1):91-115.
    [11]边肇祺,张学工.模式识别(2版)[M].北京:清华大学出版社,2001.