Temperature-based structural identification and health monitoring for long-span bridges

Temperature-based methods for the structural evaluation and monitoring of long-span bridges were investigated. The motivation for the work stemmed from the critical need for further assessment and preservation techniques for long-span bridges, which represent the most critical (and in many cases irreplaceable) nodes within the transportation network. Through this work, several methods were developed which utilize temperature as the forcing function to experimentally characterize the structure. This approach is novel and represents a potential improvement over current methods (e.g. ambient vibration monitoring) which do not allow the full transfer function of the bridge to be obtained.;In particular, this research developed and investigated three temperature-based evaluation methods. The first, termed Temperature-Based Structural Identification (TBSI), follows from the traditional structural identification framework. This approach is used for direct correlation of the input (temperature) and output responses (strains and displacements) for finite element (FE) model calibration and parameter identification. The second method, termed Temperature-Based Structural Health Monitoring (TBSHM), utilizes a streamlined approach to continually track and identify variations in key temperature-based response patterns. An interpretation framework using changes in these patterns to guide proactive maintenance and preservation practices was also developed. The last method, termed Periodic Temperature-Based Assessment (PTBA), aims to directly (i.e. without the use of an FE model or baseline information) characterize the performance of key mechanisms of a bridge by measuring physically meaningful and easily interpreted response metrics.;The research concluded temperature-based experimentation provides valuable insight into the performance of long-span bridges. TBSI exhibits accurate and reliable identification of FE model boundary and continuity conditions with clear advantages over ambient vibration model updating. The use of TBSHM has also indicated encouraging contributions to conventional SHM approaches. A distinct baseline was identified as the relationship between local strains, global displacements, and temperature variation which produces unique planes in 3D space. These 3D planes have greater sensitivity to parameter modification when compared to ambient vibration methods and show promise for identification of outliers. Additionally, quantitative performance measures were developed for PTBA of long-span bridges. This approach can improve current evaluate methods of movement mechanisms and assess long-term durability of a structure. Overall, TBSI, TBSHM, and PTBA have shown substantial benefits for advancing our understanding of constructed long-span bridge behavior.