Research On Fatigue Performance Of Reinforced Concrete Bridges Based On Measured And Extrapolated Vehicles Data

Reinforced concrete bridges are widely used in highway and railway engineering all over the world.With the accumulation of bridge service time,rapid economic growth,traffic volume increased year by year,in the role of vehicle load,many bridges in the nonreach service life on the impact of normal use of structural damage,some of which caused by fatigue.Through the research of many scholars at home and abroad,it is generally accepted that the permanent load of bridge structure is basically in a stable state within the design service life,and the change range is relatively small,and in the design process of bridge structure,there is enough safety reserve for the action of permanent load,while vehicle load is a variable load,which is dynamic and random,it is difficult to accurately consider the influence of vehicle load on bridge structure in the design process.In the previous study of traffic load,the load information is not in good agreement with the actual traffic flow,and the time of using vehicle load data is limited.In order to evaluate the fatigue behavior of reinforced concrete bridges quickly and accurately,a vehicle extrapolation model is proposed,which is in good agreement with the actual traffic flow information,the load history and stress amplitude are more consistent with the actual traffic flow.Therefore,the calculation method of extrapolating the stress amplitude of long period is put forward,which provides data support for the analysis of bridge performance.In this paper,the vehicle data of a bridge in Jingdezhen of Jiangxi Province for nearly one year are studied,and a typical t bridge with 20 m span is selected for life prediction.The major areas of work are as follows:(1)The data of vehicle type,daily traffic volume,lane distribution,vehicle spacing and vehicle weight collected by Weigh-In-Motion system in a bridge station in Jingdezhen,Jiangxi Province,are investigated and analyzed.(2)After summarizing and classifying the existing extrapolation methods,Grey theory method and BP neural network method are selected to carry out the extrapolation calculation.Using the vehicle load data obtained in Chapter 2 of this paper,based on the BP neural network theory and through the Matlab neural network toolbox,this paper adopts the vehicle parameters with different axles as variables,which is different from the traditional model,the BP neural network model for extrapolating vehicle load in short period is established and compared with the extrapolation method based on grey system theory.Using the extrapolated load data of the BP Neural Network Model,the program of simulating random vehicle flow and the program of loading the influence line of bending moment in the span of a simply supported beam are compiled with Matlab software,the load history which is more consistent with the actual traffic load is obtained.(3)Firstly,the relationship between vehicle load and fatigue life of bridge is analyzed.Based on the fatigue vehicle load spectrum of bridge obtained in Chapter 3,a typical 20 m span simply supported reinforced concrete t-shaped bridge is taken as background,the stress of bridge members is calculated under the action of extrapolation vehicle load,and then the equivalent equivalent amplitude stress amplitude is obtained.Based on the extrapolation formula of stress amplitude,the influence of vehicles with different axle number on the fatigue damage of bridge is analyzed,and the fatigue life of reinforced concrete bridge is predicted.