| 基于BP神经网络的空间索杆结构节点损伤识别研究 |
| |
| 引用本文： | 吴波,胡云霞.基于BP神经网络的空间索杆结构节点损伤识别研究[J].地震工程与工程振动,2006,26(1):83-88. |
| |
| 作者姓名： | 吴波 胡云霞 |
| |
| 作者单位： | 华南理工大学,土木工程系,广东,广州,510640 |
| |
| 基金项目： | 教育部新世纪优秀人才支持计划项目(NCET-04-0819) |
| |
| 摘 要： | 针对某实际空间索杆结构的节点损伤现象,采用BP神经网络与基于振动的损伤识别两步法对其进行了识别研究,即首先确定可能发生节点损伤的子区域,在此基础上利用对应子区域的子网络识别出具体的损伤位置和程度。识别过程中采用两个杆单元模拟发生节点损伤的杆件,用抗弯刚度降低的端部短杆单元模拟节点损伤。研究表明,虽然空间索杆结构的动力性能较为复杂,但基于结构固有频率和模态位移的组合指标对节点损伤仍较为敏感,利用它们进行节点损伤识别是有效的。
|
| 关 键 词： | 空间索杆结构 BP神经网络 节点 损伤识别 |
| 文章编号： | 1000-1301（2006）01-0083-06 |
| 收稿时间： | 2005-09-24 |
| 修稿时间： | 2005-09-242005-11-20 |
Research on damage identification of joints in a practical spatial cable-strut structure based on back-propagation neural networks |
| |
| Wu Bo,Hu Yunxia.Research on damage identification of joints in a practical spatial cable-strut structure based on back-propagation neural networks[J].Earthquake Engineering and Engineering Vibration,2006,26(1):83-88. |
| |
| Authors: | Wu Bo Hu Yunxia |
| |
| Institution: | Department of Civil Engineering, South China University of Technology, Guangzhou 510640, China |
| |
| Abstract: | Using the back-propagation neural network and the two-step damage identification method based on structural dynamic responses,joint damage in a practical spatial cable-strut structure is identified.The subregion containing damage joints is determined first,and then the damage location and the damage extent are identified in detail using sub-networks.Two elements are adopted to represent a strut with joint damage,and the shorter element with lower moment-resisting stiffness is used to simulate the joint damage.Simulation results show that the combined indexes based on natural frequencies and mode displacements are sensitive to the joint damage and effective for the(identification) of joint damage. |
| |
| Keywords: | spatial cable-strut structure back-propagation neural network joint damage identification |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
