Bridge damage detection from the equivalent damage load by multitype measurements

被引：19

作者：

Li, Yixian ^{[1
]}

Sun, Limin ^{[1
,2
]}

Zhang, Wei ^{[3
,4
]}

Nagarajaiah, Satish ^{[5
]}

机构：

[1] Tongji Univ, Dept Bridge Engn, Coll Civil Engn, Shanghai 200092, Peoples R China

[2] Tongji Univ, State Key Lab Disaster Reduct Civil Engn, Shanghai, Peoples R China

[3] Fujian Acad Bldg Res, Fuzhou, Peoples R China

[4] Fujian Key Lab Green Bldg Technol, Fuzhou, Peoples R China

[5] Rice Univ, Dept Civil & Environm Engn, Houston, TX 77005 USA

来源：

STRUCTURAL CONTROL & HEALTH MONITORING | 2021年 / 28卷 / 05期

基金：

中国国家自然科学基金;

关键词：

damage identification; equivalent damage load; inclination; load estimation; principal component analysis; strain;

D O I：

10.1002/stc.2709

中图分类号：

TU [建筑科学];

学科分类号：

0813 ;

摘要：

Traditional damage identification (DI) approaches are based on the structural modal information, which is unstable and affected by the environment. This study proposes a novel bridge DI algorithm for the elastically supported beams with a constant cross-section. The expression of the equivalent damage load (EDL) is deduced from the force-displacement relationship. The EDL only exists in the damaged areas, and it is a good damage indicator. Then, a principal component analysis-based load estimation method is adopted to estimate the external nodal force and the EDL. To reduce the influence of the measurement noise and errors, this study furtherly proposes a robust EDL-based damage indicator, which is calculated from multitype data. The numerical simulations and in-field experiments have discussed the influence of several factors, including damage scenarios and loading conditions, on the proposed damage indicator. All results have shown that the proposed damage indicator is applicable and useful. The proposed approach can detect the damage in a real-time manner with high computational efficiency, and it is useful for the in-operation bridges.

引用

页数：19