Link to Latest Report : Coming Soon
Background :
Bridge monitoring is essential for ensuring public safety and protecting public assets from failures caused by the aging and deterioration of bridge materials, as well as accidents. However, implementing monitoring technologies after bridge construction is both costly and challenging due to access limitations and factors such as theft. Traditional methodologies of bridge monitoring rely heavily on attaching sensors and/or cameras to in-service bridge elements, which particularly limit access to prestressed and reinforced concrete elements. Additionally, externally bonded sensors face durability challenges in aggressive weathering environments.
Recent catastrophic failures have underscored the urgent need for the rapid deployment of advanced bridge monitoring systems, especially for detecting accident events. Accidents involving vehicle and vessel collisions are becoming a significant concern nationwide. In Georgia, the damage to highway bridges from over-height vehicle collisions is notably increasing. Fortunately, multiple technologies are available to prevent and detect such accidents. Although it is not possible to prevent all types of collisions, particularly those resulting from negligence of collision warnings, technologies can be employed during a fabrication process to detect collisions and develop a rapid response plan. Additionally, vehicles designed for transportation and fuel efficiency, including electric/autonomous vehicles, are becoming heavier, which increases safety risks. These developments underscore the importance of timely real-world feedback and prompt responses.
Objectives :
The aim of this project is to develop affordable smart bridge elements, named Generating Artificial Intelligent Network (GAIN), which are capable of generating and analyzing bridge data through AI to enhance bridge asset management. By integrating these smart elements into strategic routes networkwide, we ultimately aim to enable gathering comprehensive empirical data to support state and national public safety, mobility, and welfare operations, including emergency evacuations and collision/fire accidents. This project mainly seeks to improve monitoring of performance metrics during fabrication, construction, and service, thereby informing maintenance, repair, and replacement (MRR) decisions. Ultimately, the goal is to empower bridge owners with data-driven insights, enhancing the quality, safety, and longevity of bridge networks.
Scope :
Task 1 – Research Available Technologies and Their Advantages and Disadvantages.
The research team will conduct a thorough investigation into the feasibility of available technologies, including strain gauges, sensors, and computer vision systems that can be embedded in prefabricated concrete and other members, within the GAIN framework. This includes examining data collected from various technologies deployed in typical bridge fabrication environments. For instance, the impact of strand layout, concrete placement, and the stressing process will be assessed to identify applicable technologies. Additionally, the durability and long-term maintenance requirements of these technologies in concrete over the service life of bridges will also be examined.
Task 2 – Design a Suitable Technological Framework for GAIN.
To create a GAIN framework, the research team will consider various aspects. For example, five independent criteria may be applied for evaluating the feasibility of GAIN elements:(1) compatibility with fabrication process; (2) construction process; (3) maintenance of wireless device; (4) power demand; and (5) in-service durability. All ideas, promising technologies, and constraints will be considered in order to tackle the problems that are defined in Task 1.
Task 3 – Build Prototype GAIN Elements
The research team will collaborate with a bridge producer to produce several low-cost, scaled-down versions and/or specialized features of the GAIN elements, all within the project budget constraints. The PIs believe this will be an iterative process in employing multiple technologies across a couple of members in order to identify best, practical, and sustainable approach to building GAIN elements such as GAIN beams.
Task 4 – Test GAIN Elements.
Empirical data collection in GAIN elements, including beams, will be tested in real-world and/or near-real-world scenarios. The research team will conduct tests on GAIN elements at its STRuctural ENGineering Testing Hub (STRENGTH) and aims to collaborate with state DOTs to deploy one of the GAIN elements in a real-world bridge structure, if opportunities permit.
Task 5 – Synthesize Data from GAIN into an Asset Management Framework.
The ‘GAIN’ framework involves AI-based empirical data analytics which will be further refined based on large datasets at the network-level. The proposed GAIN framework and associated technology will establish a foundation for quantitative decision analysis across a multitude of scenarios. Therefore, network-level AI-supported analytics will be conducted to analyze the empirical data collected from prototype GAIN members.
Research Team :
Principal Investigator : Mi Geum Chorzepa, Ph.D.
Co-Principal Investigator : Bjorn Birgisson, Ph.D.