Research On Damage Identification Of Continuous Beam Bridge Based On Moving Load Detection

With the increase of traffic loads in China,the construction speed of Highway Bridge is accelerated.Along with the high speed of the bridges construction of highway,the quality and safety requirements of bridges are constantly improved.To ensure the safety of the bridge,bridge detection is imperative.Therefore,it is an important research objective to put forward more accurate,efficient detection method based on the existing bridge detection technology.This paper aims to propose a method,which will not affect the utilization of bridge,can effectively detect the damage of continuous beam bridge using single point detection,and provide theoretical guidance for bridge detection.The displacement and acceleration response curves of three span continuous beams under the action of moving load are calculated by finite element simulation.Based on the acceleration response curves in lossless and lossy situations,the differential response curve of acceleration is calculated.Then,the acceleration differential response curve of the three-span continuous beam under the action of moving load is transformed through the wavelet change,and the damage localization of the continuous beam is preliminarily realized.Based on the wavelet transform curve,the damage localization method based on modal curvature calculation and the degree of damage degree based on artificial neural network are proposed.Finally,the damage situation of a four span continuous beam is analyzed by using the above methods to verify the accuracy of the method and the performance of the migration.The results of finite element calculation show that the acceleration response and displacement response of multi-span continuous beam under the action of moving load are insensitive to damage and cannot be directly used as an indicator of damage identification.Continuous beam single point or multi-point damage,through the determination of continuous beam acceleration response of the single point under moving load,calculating acceleration difference and wavelet transformation can realize damage localization.Wavelet transformation calculation results show that wavelet transform based on the acceleration response curve damage analysis are greatly influenced by moving load speed,when the load speed is greater than 60 km/h,positioning results there is a big error,unable to realize accurate positioning.Modal curvature calculation results show that after wavelet transform poor acceleration response curve for modal curvature calculation can accurately determine whether there is bridge damage,this method is not affected by load movement and noise.The results of artificial neural network show that the artificial neural network algorithm designed in this paper has high accuracy in calculating the damage degree of cross-continuous concrete beam bridge through modal curvature curve.