A methodology for measurement-system design combining information from static and dynamic excitations for bridge load testing. (22nd December 2019)

• Record Type: Journal Article
• Authors: Bertola, Numa J.; Smith, Ian F.C.
• Is Part Of: Journal of sound and vibration. Volume 463(2019)
• Publication Date: 2019-12-22
• Subjects: Structural identification -- Error-domain model falsification -- Optimal sensor placement -- Joint entropy -- Modal identification -- Effective independence method; Sound -- Periodicals; Vibration -- Periodicals; Son -- Périodiques; Vibration -- Périodiques; Sound; Vibration; Periodicals; Electronic journals; 620.205
• DOI: 10.1016/j.jsv.2019.114953
• Languages: English
• ISSNs: 0022-460X

Abstract

Abstract: Managing infrastructure assets is challenging for developed countries because of demand for increases in capacity, the scarcity of economic and environmental resources as well as ageing. Due to conservative approaches to construction design and practice, infrastructure often has hidden reserve capacity and its estimation may improve asset-management decisions. Static and dynamic bridge load testing has the potential to support engineers in their evaluation of infrastructure reserve capacity if monitoring data are associated with a robust structural-identification methodology. As choices of sensor types and locations directly influence structural-identification outcomes, sensor-placement methodologies have recently been developed to ensure successful model-updating results. Due to the nature of static and dynamic measurements, sensor-placement methodologies are usually developed independently. However, both types of load testing are used to update the same bridge behavior model. Therefore, when sensor-placement strategies are established independently, redundant sensor information is likely. In this study, two measurement-system design methodologies are proposed. First, a new methodology for sensor-placement for dynamic load testing is presented, where expected information gain of natural frequencies is used to prioritize sensor-location selections. Then, a measurement-system-design methodology combining information of both static and dynamic load testing is proposed. Finally, the methodology is evaluated using a full-scale bridge. A well-designed measurement system based on expected information gain enhances system identification and reserve-capacity estimation.