Towards Smart and Cost-effective Bridge Infrastructure Monitoring Systems
Abstract
Bridge health monitoring (BHM) has recently gained significant interest worldwide in the inspection and maintenance of aging bridge infrastructure in the era of climate change and adverse weather conditions. However, extensive datasets resulting from these monitoring systems require appropriate tools to diagnose the data systematically under various operating conditions of bridges, leading to expensive and time intensive BHM strategies. To mitigate this challenge, a smart and cost-effective bridge infrastructure management system is of paramount need in today’s world. This thesis aims to develop a suite of cost-effective bridge management strategies by employing limited and mobile sensing technology and addressing their inherent challenges in real-world situations. First, a limited sensor-based cost-effective approach is developed to analyze the traffic-induced nonstationary vibration response of the bridge. The proposed technique can deal with practical challenges of direct BHM, such as traffic interruptions, bridge closures, limited space, and the limited number of sensors, thereby eliminating the need for high labor and equipment costs. Secondly, the visualization of BHM data is explored for systematic diagnosis of the bridge data. A visualization tool based on Bridge Information Modeling (BrIM) is proposed which is suitable for real-time system identification of bridges. The objective of the proposed tool is to take one step forward from static to dynamic BrIM by representing and visualizing real-time BHM data.
Contact-based BHM usually involves direct instrumentation with sensors to extract the modal parameters from the ambient or forced vibrations. As an alternative to direct BHM, indirect BHM (iBHM) has emerged as a promising avenue for effective and inexpensive monitoring of bridge infrastructure. However, the existing iBHM methods face challenges associated with the accurate identification of bridge properties under various driving and vehicle conditions. In this thesis, a hybrid time-frequency method is proposed for decoupling vehicle bridge interactions and performing robust bridge modal identification under various operational challenges. The method is capable of bridge condition assessment using vehicle response from a passing vehicle traveling over a bridge, resulting in a smart drive-by BHM technology. The vehicle response in iBHM is often criticized as the presence of vehicle frequency can make vehicle scanning ineffective. Therefore, this thesis also explores the robust contact point (CP)-based BHM method, which is free from vehicle conditions and provides more accurate estimates of bridge frequencies.