用于结构损伤位置识别的两种神经网络性能比较
|
摘要
为了克服用BP网络进行结构损伤位置识别时网络结构确定难、网络训练易陷入局部极小、训练时间长以及处理带噪声数据需要大量的训练样本等问题,提出用SOFM网络进行结构损伤位置识别的方法。分别用SOFM网络和BP网络对一桁架结构进行损伤位置识别,通过比较两种网络的性能发现SOFM网络不但网络结构容易确定,网络训练不存在陷入局部极小的问题,BP网络只有在大量训练样本条件下才能保证网络具有较好的抗噪声能力,若训练样本不足,则BP网络的抗噪声性能较差,而SOFM网络在较少训练样本情况下即可具有良好的抗噪声性能,因而SOFM网络更适合训练样本有限条件下的结构损伤位置识别。
BP neural networks used in the structural damage location detection have many shortcomings,such as difficult network structure deciding,training time consuming and easily trapped into local minimum,and a lot of training samples needed to identify the noise contaminated data.To overcome these shortcomings,a method using SOFM neural network to identify the damage location is put forward.The two types of networks are used to identify the damage location of a truss structure.The comparison of the performance between the two types of neural network shows that the structure of the SOFM neural network is easier to be decided and its training won't be trapped into local minimum,it has better antinoise capability even the training samples are limited,while the BP network can have good antinoise capability only when there are plenty of training samples,otherwise the BP network will have poor antinoise capability.It is concluded that the SOFM neural network is better than the BP network to identify the damage location when the training samples are limited.
BP neural networks used in the structural damage location detection have many shortcomings,such as difficult network structure deciding,training time consuming and easily trapped into local minimum,and a lot of training samples needed to identify the noise contaminated data.To overcome these shortcomings,a method using SOFM neural network to identify the damage location is put forward.The two types of networks are used to identify the damage location of a truss structure.The comparison of the performance between the two types of neural network shows that the structure of the SOFM neural network is easier to be decided and its training won't be trapped into local minimum,it has better antinoise capability even the training samples are limited,while the BP network can have good antinoise capability only when there are plenty of training samples,otherwise the BP network will have poor antinoise capability.It is concluded that the SOFM neural network is better than the BP network to identify the damage location when the training samples are limited.
引文
[1]王术新,姜哲.基于结构振动损伤识别技术的研究现状及进展[J].振动与冲击,2004,23(4):99-102.
[2]谢强,薛松涛.土木工程结构健康监测的研究状况与进展[J].中国科学基金,2001(5):285-288.
[3]黄光远,刘小军.数学物理反问题[M].济南:山东科学技术出版社,1993.
[4]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.
[5]WORDEN K,BALL A,TOMLINSON G.Neural networks for fault loca-tion[C]//Proc.of the 11thInternational Modal Analysis Confer-ence.1993:47-54.
[6]TSOU P,SHEN M H H.Structural damage detection and identificationusing neural networks[J].AIAA Journal,1994,32(1):176-183.
[7]YEUNG W T,SMITH J W.Damage detection in bridges using neuralnetworks for pattern recognition of vibration signatures[J].EngineeringStructures,2005,27(5):555-578.
[8]王柏生,倪一清,高赞明.用概率神经网络进行结构损伤位置识别[J].振动工程学报,2001,14(1):60-64.
[9]魏海坤.神经网络结构设计的理论与方法[M].北京:国防工业出版社,2005.
[10]王广军,陈红,何祖威,等.锅炉运行模式识别方法及应用[J].中国电机工程学报,2003,23(6):204-208.
[11]张立明.人工神经网络的模型及其应用[M].上海:同济大学出版社,1993:141.
[12]GIRAUDEL J L,LEK S.A comparison of self-organizing map algorithmand some conventional statistical methods for ecological community ordi-nation[J].Ecological Modelling,2001,146:329-339.
[13]NI Y Q,WANG B S,KO J M.Selection of input to neural networks forstructural damage identification[J].Smart Systems for Bridges,Struc-tures and Highways,Proceedings of The International Society for OpticalEngineering,1999,3671:270-280.
[14]CYBENKO G.Approximation by Superposition of Sigmoid Function[J].Mathematics of Control.Signals and System,1989(2):304-314.
[15]丁幼亮,李爱群,缪长青.大跨斜拉桥扁平钢箱梁损伤定位指标的噪声鲁棒性比较研究[J].公路交通科技,2007,24(4):79-83.
[2]谢强,薛松涛.土木工程结构健康监测的研究状况与进展[J].中国科学基金,2001(5):285-288.
[3]黄光远,刘小军.数学物理反问题[M].济南:山东科学技术出版社,1993.
[4]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.
[5]WORDEN K,BALL A,TOMLINSON G.Neural networks for fault loca-tion[C]//Proc.of the 11thInternational Modal Analysis Confer-ence.1993:47-54.
[6]TSOU P,SHEN M H H.Structural damage detection and identificationusing neural networks[J].AIAA Journal,1994,32(1):176-183.
[7]YEUNG W T,SMITH J W.Damage detection in bridges using neuralnetworks for pattern recognition of vibration signatures[J].EngineeringStructures,2005,27(5):555-578.
[8]王柏生,倪一清,高赞明.用概率神经网络进行结构损伤位置识别[J].振动工程学报,2001,14(1):60-64.
[9]魏海坤.神经网络结构设计的理论与方法[M].北京:国防工业出版社,2005.
[10]王广军,陈红,何祖威,等.锅炉运行模式识别方法及应用[J].中国电机工程学报,2003,23(6):204-208.
[11]张立明.人工神经网络的模型及其应用[M].上海:同济大学出版社,1993:141.
[12]GIRAUDEL J L,LEK S.A comparison of self-organizing map algorithmand some conventional statistical methods for ecological community ordi-nation[J].Ecological Modelling,2001,146:329-339.
[13]NI Y Q,WANG B S,KO J M.Selection of input to neural networks forstructural damage identification[J].Smart Systems for Bridges,Struc-tures and Highways,Proceedings of The International Society for OpticalEngineering,1999,3671:270-280.
[14]CYBENKO G.Approximation by Superposition of Sigmoid Function[J].Mathematics of Control.Signals and System,1989(2):304-314.
[15]丁幼亮,李爱群,缪长青.大跨斜拉桥扁平钢箱梁损伤定位指标的噪声鲁棒性比较研究[J].公路交通科技,2007,24(4):79-83.
版权所有：© 2023 中国地质图书馆 中国地质调查局地学文献中心