Study Of Weight Limit And Vehicular Overloading Induced Fatigue Damage For Steel-concrete Composite Girder Bridges

Vehicular overloading could induce serious accumulated fatigue damage in bridges,accelerate the deterioration of bridge components,and even lead to bridge collapse.Rationally and accurately estimating the cumulative fatigue damage of the bridge under overloading is essential for the bridge fatigue design and safety evaluation.In this study,the cumulative fatigue damage of the small-to-medium-span steel-concrete composite girder bridge was analyzed systematically and the corresponding vehicle weight limit methods were developed.More specific research work was conducted with focus on the following aspects:(1)The fatigue design models specified in the bridge design codes for Europe,the United States,the United Kingdom,and China were reviewed and analyzed systematically in terms of their development and corresponding adoptions in fatigue design.Since the real traffic load is very sophisticated and can vary dramtically from site to site,the fatigue design models differ from code to code.The development of the fatigue design model ? specified in the Chinese code learned much from the configurations and loading conditions for the fatigue deisgn models specified in the European code and the British code.Since the actural truck traffic in China is different with that in Europea,it is highly desirable to adjust the fatigue design models specified in the Chinese code based on the consideration of Chinese traffic conditions in order to evaluate the fatigue damage of bridges in China more accurately.(2)A typical steel-concrete composite girder bridge was employed to investigate the coupled corrosion-overloading effect on the fatigue life of bridges.The corrosion-overloading coupling process was realized by adopting the deteriorated cross section of the steel gird er due to corrosion when calculating the bridge responses under the action of truck overloading.A simple corrosion fatigue estimation method was proposed by considering the coupled corrosion-overloading effect,which consists of the individual effects due to pure overloading,pure corrosion,and the corrosion-overloading interaction.The results showed that the coupled corrosion-overloading effect can greatly reduce the fatigue life of bridges and should be carefully considered in bridge fatigue design whe rever necessary.(3)A method for determining the vehicle weight limit and overload permit checking was proposed based on the consideration of the cumulative fatigue damage of bridges.A typical steel-concrete composite girder bridge was used as an example for illustrating the proposed method.Based on the results from this study,the rationality of the federal bridge formula,i.e.,Bridge Formula B,was discussed.The results from this study can not only be used to develop vehicle weight limits and assist in overload permit checking but also to assess the fatigue damage and predict the remaining fatigue life of existing bridges.(4)In this part,the fatigue damage of bridge system under truck overloading was investigated using the open source framework OpenSEES.The bridge system was regarded as a series-parallel system consisting of the bridge deck slab sub-system and the steel girder sub-system,and the fatigue damage of the bridge was analyzed parametrically based on the behavior of components and that of the system.Finally,a method for determining rational vehicle weight limits for the highway bridge was proposed considering the fatigue damage of the bridge system.The proposed method could also be used for estimating the fatigue damage of bridges due to truck overloading.(5)The prediction of fatigue failure probability of bridges due to traffic overloading was investigated by using the machine-learning-based method.This method was proposed for the purpose of replacing the t raditional approaches,such as the finite element analysis and reliability analysis.The whole procedure consists of three parts,namely,the generation of training data based on the finite element analysis,the development,train,validation,and test of the artificial neural network,and the prediction of fatigue failure probability for the bridge.The comparation of the results obtained from the finite element analysis and the machine-learning-based method was conducted,and it was found that their difference was not significant.Results showed that based on a limited set of training data for the bridge under consideration,the proposed machine-learning-based approach can assist in providing an instantaneous assessment of the fatigue failure probability for bridges with high accuracy.