Statement of Problem
As aging bridges in the United States undergo wear-and-tear deterioration, bridge owners must expend significant resources to vigilantly inspect and rehabilitate their inventory. In recent years, this task has become increasingly challenging due to reductions in the economic resources necessary to fund these efforts. The need to invest in bridge inspection and repair is necessary to avoid catastrophic failure of these critical infrastructure elements. The loss of human life and the long-term economic impact of a failed bridge can be enormous as was the case in Minnesota when the I-35 Bridge collapsed in 2007. Even the partial failure of a critical bridge component can represent an expensive management issue that can adversely affect bridge users for a long period of time. For example, the San Francisco-Oakland Bay Bridge experienced a failed eyebar in 2009. After detection, the bridge was closed for a short-period followed by nightly lane closures for 5 weeks as construction crews repaired the element. The issue also resulted in more frequent visual inspections of the bridge eyebars. The current approach to bridge inspection largely relies on visual inspections conducted by professionally trained inspectors with years of experience. While this approach to bridge management has served the bridge engineering community well for many years, it is potentially not a scalable management approach as the inventory of bridges rapidly ages. In addition, the information obtained by visual inspections can contain a high degree of variability due to the subjectivity of the inspectors. In some cases, inspectors are required to conduct their visual inspection in unpleasant work environments that make the inspection process more challenging.
Wireless Bridge Monitoring System to Long-Span Bridges
The emergence of low-cost sensors and data acquisition systems has now made permanently installed monitoring systems a possibility for bridges. The availability of real-time data from a bridge monitoring system can aid bridge owners in more objectively evaluating the conditions of their structures. The vast majority of monitoring systems deployed on operational bridges have been based on the use of tethered (i.e., wired) system architectures. However, the installation of tethered sensors, specifically their extensive wiring, can be costly and labor intensive. In addition, the coaxial wires can be vulnerable to physical failure. In response to these limitations, wireless communication technologies have been proposed for future bridge monitoring systems. Wireless bridge monitoring systems provide bridge owners with the option of making a smaller initial investment (largely because wires have been eradicated) while still offering the benefits gained from long-term bridge monitoring. While wireless bridge monitoring systems have shown great promise, their long-term reliability in real bridges is not yet well explored with only a limited number of efforts reported in the literature. Other major impediment in applying wireless sensors in long-span bridge systems is the lack of viable power sources that can sustain operations for long periods of time (e.g., years, decades) without requiring the physical replacement of batteries. The continuous operation of a wireless bridge monitoring system requires the use of low-power hardware components coupled with an appropriate power harvesting technology (e.g., solar cells, miniature wind turbines, vibration power harvester).
Step-Wise Data Management Service via Hierarchically Designed Cyberinfrastructure
A wireless structural health monitoring system designed for long-span bridges should be autonomous and not require regular maintenance. In that spirit, this study explores the creation of hardened wireless monitoring system that can operate for long periods of time without human intervention. To provide the system with complete autonomy and to ensure the system is continuously checked for component failures, the wireless monitoring system installed in a bridge is remotely controlled by cyber-environment that manages the flow of data and information from the sensor to the end-user. The cyber-environment is hierarchically designed using two major tiers (Figure). In the left tier, a low-power wireless sensor network constructed from the Narada wireless sensing unit is deployed within a bridge to collect bridge responses and environmental data continuously, on a schedule or on demand by system end users. Once data is collected, the Narada sensor network processes the raw sensor data using in-network data interrogation methods in order to consolidate it into a compact format. The Narada nodes then communicate that data through an on-site server to the right tier with internet-enabled cyberinfrastrucure via a third generation (3G) cellular network connection. The hieratical placement of two tiers essentially wraps the physical wireless sensor system into the complete cyber-environment. A robust middleware with a set of server and client application interfaces supports two-way communication between the on-site Narada server and servers remotely located on the internet including a database server, an application server and a remote terminal server. The hierarchical design of the proposed wireless bridge monitoring system enables online system diagnosis (i.e., to ensure the long-term durability of the monitoring system) and system reconfiguration (e.g., change sample rate, modify collection schedule).
Implementation of Proposed System to New Carquinez Suspension Bridge
To create a wireless monitoring system with enhanced longevity in the field, a power management system based on solar energy is adopted. An assessment of the long-term performance of the proposed wireless monitoring system is ongoing at the New Carquinez Bridge (NCB). A permanent wireless monitoring system is installed consisting of 21 wireless sensor nodes collecting data from over 65 channels. The early efforts have included performance verification of a dense network of wireless sensors installed on the bridge and the establishment of a cellular gateway to the system for remote access from the internet. Acceleration of the main bridge span was the primary focus of the initial field deployment of the wireless monitoring system. An additional focus of the study is on ensuring wireless sensors can survive for long periods without human intervention. Toward this end, the life-expectancy of the wireless sensors has been enhanced by embedding efficient power management schemes in the sensors while integrating solar panels for power harvesting. To illustrate the ability to autonomously process the bridge response data, stochastic subspace identification (SSI) method is used to extract accurate modal information of the bridge. The dynamic characteristics of the NCB under daily traffic and wind loads have been compared to a high-fidelity finite element model of the bridge. The Internet-enabled wireless structural monitoring system proved to be scalable to a large number of nodes, reliable for long-term use and provided highly accurate bridge response and environment data.