Collapse of the I-35W Bridge in Minnesota has resulted in considerable interest in steel truss and gusset plate connection performance. The load paths in many of these bridges are non-redundant and thus failure of a truss member or connection may cause collapse of the structure. Periodic inspections and structural evaluations are crucial for these types of bridges. The most common method of evaluation that has been used to assess the safety of highway bridges is load rating, an approach used to estimate the available strength and allowable load on a bridge. Bridge load rating computer programs do not explicitly consider the gusset plates connecting the truss members. Hence, after the initial design calculations are completed and checked, it is unlikely that recalculations for load rating purposes have been made for gusset plates. As an outcome of the investigation into the collapse of the I-35W Bridge, steel truss bridge connections are required to undergo review. This additional scrutiny requires development of new tools to efficiently and effectively evaluate the large numbers of steel truss bridge connections in the inventory. Digital imaging techniques have been developed at Oregon State University to enable rapid collection of field geometric data from in-service gusset plates. These tools are implemented in software that allows extraction of gusset plate dimensional information to facilitate ratings. The present tools provide a basic set of functionality including rectification, metrification, and selected data extraction such as length, perimeter, and angles. These basic functions need enhancement to take full advantage of the advancements available to bridge inspection and management with digital imaging. Such enhancements are proposed to enable transportation agencies to efficiently and effectively collect geometric and condition data and use this data to evaluate and rate gusset plate connections.

There are four main objectives of the proposed research: 1. Develop methods to collect dimensional gusset plate connection information including surface geometry and out-of-plane deformations on in-service gusset plates. 2. Develop methods to automate identification and optimization of reference target points. Develop methods to automate extraction of gusset plate edge locations, fastener locations and their corresponding member affiliations, and member orientations. 3. Develop finite element modeling and analysis techniques to directly rate gusset plates using extracted digital image data as the input source. 4. Develop software tools to manage and organize images and image data to enhance bridge management and allow identification of condition changes over time.

Task 1: Literature Review: Review technical literature on image analysis, machine vision, and finite element analysis applicable to steel gusset plate connection evaluation. Task 2: Software Development and Data Collection Develop enhanced procedures for field inspectors to collect high fidelity three-dimensional geometric information of gusset plates. These procedures must be applicable to commonly encountered truss joint configurations. Execution of the procedures will involve the use of image editing software that will be developed through this research to allow for correction of out-of-plane distortion, image resizing and scale calibration, and extraction of dimensional measurements from the images. Task 3: Gusset Plate Finite Element Analysis: Develop a finite element analysis modeling and analysis protocol that can directly rate gusset plate connections for allowable stress rating, load factor rating, or load and resistance factored rating from the metric digital orthographs. Alternatively, inputs can be provided by the user from available drawings or other sources. Task 4: Implementation Example: Develop an example gusset plate analysis demonstrating the field data collection methods. The plate analysis will be benchmarked with ABAQUS and also compared with traditional design methods and the FHWA Guidelines for gusset plate connection evaluation. Task 5: Imaging Data Informatics for Bridge Management: Develop methods to manage, categorize, organize, and query digital orthographs to enable mapping correspondences with bridge inspection records and enhance long-term bridge management. Database structures will be developed that will enable image comparisons to be made between inspection intervals. Multi-scale imaging methods will enable coarse and fine metric features to be mapped and referenced to each other. A full set of truss bridge metric images will be collected, rectified, categorized, mapped, and linked to available drawings and inspection notes. Requirements of the imaging system will be developed and implementation protocols developed.

The project has a well developed work plan. Potential partners should review the work plan with the understanding that while there will be some flexibility in tailoring this work to the needs of partners, the scope of this project is substantially set.