桥梁健康监测中桥梁结构损伤智能识别方法研究  

Research on Intelligent Identification Method of Bridge Structure Damage in Bridge Health Monitoring

作　　者：蒋红[1] JIANG Hong(Anhui Technical College of Water Resources and Hydroel ectric Power,Hefei 231603,China)

机构地区：[1]安徽水利水电职业技术学院,安徽合肥231603

出　　处：《佳木斯大学学报（自然科学版）》2023年第6期118-121,共4页Journal of Jiamusi University：Natural Science Edition

基　　金：2023年度高校科学研究项目(自然科学类)(2023AH053048)。

摘　　要：作为桥梁健康监测中的重要组成部分,结构损伤识别对桥梁维修和管理有着重要意义。为优化桥梁结构损伤智能识别效果,研究提出了基于小波变换改进和深度置信网络的桥梁结构损伤智能识别方法。结果显示,该方法对斜拉索受损模式的最大识别准确性可达到0.95,相较于NB算法提高了0.12。这表明,该方法增强了对复杂桥梁数据的处理能力,提高了桥梁结构损伤智能识别的准确性,有利于促进桥梁结构损伤识别技术的发展,为提升桥梁管理水平提供保障。As an important part of bridge health monitoring,structural damage identification is of great significance to bridge maintenance and management.In order to optimize the intelligent identification effect of bridge structure damage,an intelligent identification method of bridge structure damage based on wavelet transform improvement and depth confidence network is proposed.The results show that the maximum recognition accuracy of the method for damage modes of stay cables can reach 0.95,which is 0.12 higher than that of NB algorithm.This shows that this method enhances the processing ability of complex bridge data,improves the accuracy of intelligent identification of bridge structure damage,is conducive to promoting the development of bridge structure damage identification technology,and provides guarantee for improving the level of bridge management.

关 键 词：桥梁结构 损伤 深度置信网络 小波变换 

分 类 号：U447[建筑科学—桥梁与隧道工程]