基于谱系聚类分析的桥梁结构模态参数自动化识别方法研究

发布时间：2018-09-06 07:41

【摘要】：桥梁结构模态参数识别作为桥梁健康监测系统的重要组成部分之一,对桥梁的长期健康监测具有重要意义。而现有的模态参数识别算法还不能实现结构系统阶次和模态参数的自动化识别,基于此,简单介绍了随机子空间算法的基本原理;通过引入"滑窗技术"以实现对时变结构模态参数的在线识别;针对稳定图定阶难这一问题,提出了基于奇异熵理论的系统定阶算法,以实现系统阶次的自动化确定;又将统计学中的"谱系聚类算法"与随机子空间算法进行结合,实现桥梁结构模态参数的自动化识别。利用该算法识别苏通大桥在竖桥向和横桥向的模态参数结果,并将其与理论计算值进行对比分析,结果表明:算法不仅能实现桥梁结构系统阶次的自动化确定,还能实现模态参数的自动化识别,方便对桥梁结构参数的实时监控,了解桥梁结构的健康运营状态。
[Abstract]:As an important part of bridge health monitoring system, bridge modal parameter identification is of great significance to bridge long-term health monitoring. However, the existing modal parameter identification algorithms can not realize automatic identification of structural system order and modal parameters. Based on this, the basic principle of stochastic subspace algorithm is introduced briefly. In order to solve the problem that it is difficult to determine the order of the stability diagram, a system order determination algorithm based on the singular entropy theory is proposed to realize the automatic determination of the order of the system. Automatic identification of modal parameters of beam structure. This algorithm is used to identify modal parameters of Sutong Bridge in vertical and transverse directions, and the results are compared with the theoretical values. The results show that the algorithm can not only realize automatic determination of the order of bridge structure system, but also realize automatic identification of modal parameters. The real-time monitoring of beam structure parameters can help to understand the healthy operation state of bridge structures.
【作者单位】：重庆交通大学土木工程学院;南充市住房和城乡建设局;西南交通大学土木工程学院;
【基金】：国家自然科学基金(11627802;51478071)
【分类号】：U446

【相似文献】