Couto Braguim, Thales. A framework for the assessment of the remaining fatigue life of steel bridges with welded joints. Retrieved from https://doi.org/doi:10.7282/t3-y00s-w330
DescriptionIn the United States, bridge owners rely on the AASHTO Manual for Bridge Evaluation (MBE) to estimate the remaining fatigue life of steel bridges. Although the manual allows refinements on the fatigue load model by including data from field measurements, the fatigue resistance model, which is based on the AASHTO LRFD S-N curves, has been shown to overestimate fatigue damage. Since the outcomes of the remaining fatigue life assessment may result in costly actions as repairing or replacing the deficient member, refinements on fatigue resistance models are desirable and could benefit bridge owners in making better decisions. Another area in the fatigue evaluation process is to consider the future impact of truck platoons on existing bridges. A framework for applying deterministic as well as probabilistic fatigue assessment of welded joints is presented in this research considering three different approaches:1) stress-life, 2) Linear Elastic Fracture Mechanics (LEFM), and 3) the UniGrow approaches. The current AASHTO MBE methodology was used as the benchmark model. The stress-life fatigue resistance model was developed based on the AASHTO fatigue database by refining and reformulating the current S-N curves as bilinear S-N curves for the fatigue detail categories E and E’. Due to the great variability of fatigue lives in the LEFM models caused by different initial crack sizes, this research proposes using the LEFM coupled with the Phased Array Ultrasonic Testing (PAUT) technique. The PAUT has enhanced capabilities of detecting flaws in comparison with other Non-Destructive Testing (NDT) methods. This research also applied the Unigrow model, which is based on the strain-life method and is capable of including the residual stress effects in the fatigue analysis. It was found that the remaining fatigue life obtained from the UniGrow model and from the LEFM coupled with PAUT have similar values. The proposed resistance models were applied to the welded joints of a steel bridge used as a case study. The fatigue loading for the case study was based on a calibrated Finite Element Model integrated with site-specific weigh-in-motion (WIM) measurements.
Additionally, in order to address the impact of truck platooning in the fatigue behavior of steel bridges, this research investigated the cumulative fatigue damage caused by truck platoons made of two, three, and four trucks. It was found that although the load effect caused by truck platoons is higher than a single truck, the number of fatigue stress cycles is reduced in certain cases. Therefore, for some cases, a truck within a platoon causes less fatigue damage than a single truck.