Book Description

Structural health monitoring (SHM) has drawn significant attention in recent decades because of its potential to reduce maintenance costs and increase the reliability of structures. An important class of structures that can potentially benefit from SHM are bridges, many of which are structurally deficient due to lack of adequate maintenance. Through condition assessment of these bridges, an effective plan of maintenance can be determined, offering the possibility to prolong service life, as well as to prevent catastrophic disasters due to sudden collapse. To date, numerous damage detection algorithms have been proposed. Still, challenges remain in applying such algorithms to monitor bridges in the field. In reality, the extent of an SHM system is limited by available budgets, which define the number of sensors that can be deployed. This dissertation first presents a damage detection algorithm using static strain developed for efficient structural condition assessment with a few sensor nodes. A laboratory moving vehicle experiment has been developed for validation of the approach. However, just a few sensor nodes in SHM system cannot provide detailed information on damage location. A solution to include many sensors within a limited budget with increased efficiency is to use a Wireless Smart Sensor Network (WSSN) because of the merits of low cost, easy installation, and effective data management. An acceleration-based SHM algorithm for WSSN has been developed with a decentralized network topology. This approach has been implemented into a modularized damage detection service. The SHM application is designed to leverage the on-board computation capability of the WSSN, reducing the transmitted data size by distributing the computation burden. The SHM application for WSSN has been validated in lab-scale experiments on a truss bridge model. Nonetheless, the real challenge of SHM is in the deployment on full-scale bridges for continuous monitoring. The usability and stability of WSSN has been validated on an architectural staircase in the Siebel Center. Based on the usability investigation, the deployment of the world0́9s largest WSSN on the Jindo Bridge, a cable-stayed bridge has been achieved in South Korea. The main purpose of the deployment was to validate the bridge monitoring system using WSSN and energy harvesting devices in a long-term manner. The ultimate goal of this dissertation is to deploy the developed on-board decentralized damage identification application using WSSN on a historic truss bridge. As a first step, a series of dynamic tests were conducted for modal analysis using both wired and wireless sensor systems. During the tests, the functionality of the wireless sensor system with ISHMP Services Toolsuite was confirmed. For model-based damage identification approach developed herein, a finite element (FE) model was created. The initial FE model was updated based on a visual estimate of the corrosion. The updated model was used to generate baseline information for damage detection. Finally, the WSSN-based autonomous SHM system using the decentralized damage detection application was deployed on the historic bridge. The permanent SHM system was installed on the bridge, and the damage detection application was successfully run on the bridge. The damage detection results using the comprehensive application will be compared with those from the measured data. In summary, this dissertation provides a robust SHM system for bridge structures in use of WSSN. The decentralized damage detection approach is experimentally validated for WSSN. The performance of WSSN and energy harvesting devices will be evaluated.