ABSTRACT
ABSTRACT: Bridge load rating is commonly performed to assess the level of load carrying capacity of a bridge under standard loads. Load rating process involves a certain level of analysis of the bridge structure commonly using nominal sectional properties. However, on site conditions are generally different than those listed in the blue-prints and variations in structural properties such as support conditions, section or composite action losses, deterioration due to aging, salting, corrosion etc should also be considered for load rating of a bridge. One of the best ways to conduct a load rating that would consider current condition of a bridge involves non-destructive testing and structural identification based approach. In this paper, a two dimensional grid model updating approach based on field measured data and related rating studies are explained. The proposed grid model has advantages of better simulating bridge geometry and loading conditions as compared to the 1D beam models. Model updating or calibration process is not only based on collected dynamic data (mode shapes, modal frequencies, and order of modes) but also utilizes static deformed shape information (BGCI). Usage of deformed shape and dynamic data together imposes advantages on structural identification. Optimization routines for calibration are also improved by grouping variables at different stages of calibration improving the chance and speed of finding the global minimum of objective function. Usage of 2D grid based modeling, calibration, and rating program developed under MATLAB language is demonstrated using a concrete-deck-on-steel-stringer example bridge.
