|
| Experimental and numerical studies on precise damage localization of a long-span steel truss bridge model |
| Received:October 08, 2010 Revised:December 28, 2010 |
| View Full Text View/Add Comment Download reader |
| DOI：10.7511/jslx201105014 |
| KeyWord:precise damage localization damage identification partial nodes stochastic damage locating vector method steel truss bridges |
| Author | Institution |
| 安永辉 |
大连理工大学 建设工程学部 土木工程学院,大连 |
| 欧进萍 |
大连理工大学 建设工程学部 土木工程学院,大连 |
|
| Hits: 2272 |
| Download times: 1578 |
| Abstract: |
| Large numbers of long-span steel truss bridges with long bars are in service. Therefore, it is imperative for their safety assessment and maintenance to obtain damaged bars and the location of local damage. To solve this problem, a simply supported bailey steel truss bridge Benchmark model for health monitoring was designed and constructed according to the highway and railway bi-purpose steel truss bridges employed in China. Based on structure form and detailed dimensions, a Matlab finite element model was established as the reference model due to the perfect match of the modal parameters with experimental results. Under ambient excitation, a triangular substructure was separated as a research object. Firstly, based on the Stochastic Damage Locating Vector (SDLV) method, the damaged bar was detected and additional sensors were installed on it to identify the local damage. Secondly, conventional SDLV method needs information from all nodes in a truss structure or a substructure of it, which results in high cost. Possibility of damage localization based on data from partial nodes in a substructure was discussed preliminarily. Both the numerical and experimental results indicate that the SDLV method can not only identify the damaged bars but also precisely locate the local damage; moreover, which lays a good foundation for damage detection based on the SDLV method using partial nodes. |
|
|
|