Developing Fatigue Vehicle Models for Bridge Fatigue Assessment under Different Traffic Conditions

The fatigue damage of steel bridges induced by heavy traffic loads is a critical problem worldwide. A uniform fatigue vehicle is generally adopted in bridge design codes to evaluate the cumulative fatigue damage of steel bridges. However, since the traffic loads vary across sites, a single vehicle with predetermined parameters may be imprecise for actual fatigue damage estimations. In this study, a new method for developing fatigue vehicle models that are applicable to various traffic conditions is proposed. The traffic in China was taken as an example for illustrating the proposed method. Numerical simulations were performed to investigate the fatigue damage distribution at 12 typical weigh-in-motion sites in China based on the collected vehicle load spectra, and the errors in fatigue damage estimations when adopting the fatigue vehicle Model III in Chinese bridge code were evaluated under various scenarios. Results showed that using Model III for fatigue analysis may seriously underestimate or overestimate the cumulative fatigue damage caused by the actual traffic loads under some conditions. Using a modified Model III with gross weight adjusted to the site-specific traffic condition could lead to a significant improvement, but the errors are still within a relatively large range. Following the classification and mathematical optimization techniques of the proposed method, a three-axle fatigue vehicle model and a four-axle fatigue vehicle model corresponding to the light and heavy traffic, respectively, were developed. These two fatigue vehicle models were proven to produce better and more consistent accuracy than the single fatigue vehicle model (i.e., Model III) for various traffic compositions in China.

Automated summary

This study proposes a new method for developing fatigue vehicle models applicable to various traffic conditions, illustrated through a case study of traffic in China, showing the errors in estimated fatigue damage using the existing fatigue vehicle model under actual traffic loads, and introducing an improved model for more accurate fatigue damage estimation.