Study On Suspender-replacement Optimization Of Tied Arch Bridge

The tie-bar arch bridge has developed rapidly in China in recent decades and has become one of the important bridge types in China.However,due to insufficient design considerations and post-maintenance,the damage of the suspender is broken,resulting in a large number of tied arch bridges.Need to replace the suspender.In this paper,based on the cable-changing project of a tied arch bridge in Dongguan,the cable force value is calculated by the vibration frequency method based on the self-vibration frequency,and the cable force value is calculated by comparing the various cable-measuring methods.Meet the engineering error requirements and provide accurate cable force values for subsequent optimizations.Based on the review and analysis of a large number of bridge replacement data and BP neural network,this paper calculates the bridge deck elevation and phase during the replacement process for the complex bridge structure of the tied arch bridge.The change of the cable force of the adjacent suspender is compared and verified by the results of the unoptimized scheme,the optimization scheme,the finite element calculation and the BP algorithm.Finally,the optimal design scheme of the equivalent lift of the secondary cable force is proposed.It provides an effective basis for the construction of the reinforcement project and the control of the deck deformation.Finally,a safe and effective scheme is selected.Using Midas finite element analysis software as a tool,the modeling and calculation of the tied arch bridge provides training samples and parameters of the optimization scheme for the BP network.The network is tested by the verification sample,and the calculated value of the network and the expected output of the test sample are tested.Values are compared to verify the correctness and validity of the network.The cable-changing project adopts the optimized design scheme,and takes the cable force change value of each tension and unloading as the input value of the BP network,and compares the measured data with the network-calculated predicted data to prove the effectiveness of BP neural network in this project.