Study On Finite Element Model Updating Of Bridge Structure Based On Health Monitoring

Bridge engineering is the national lifeline project.After the bridge opened to traffic operation,with the passage of time,it will be subject to environmental erosion,human factors,material aging,natural disasters and vehicle load interaction,which subject to varying degrees of damage and deterioration.Bridge health monitoring system structural response was obtained from the sensors in real-time monitoring data information and the environmental conditions and bridge of data calculation and analysis,judgment and evaluation of bridge structure stress and resistance attenuation law,ensuring the sound operation of the bridge.The core of the health monitoring system is the damage identification,however,damage identification is based on the establishment of accurate finite element model.In this paper,the dynamic response monitored by the bridge health monitoring system is the research target of the finite element model,with neural network algorithm and the response surface method in the application of the model updating as the research object.The finite element model was analyzed by ANSYS software.Optimization of Neural Network by Genetic Algorithm,using MATAB programming algorithm to achieve model updating.At the same time,using the MATAB developed toolbox,realized the visualization of the response surface model modification method.Based on the health monitoring system for real-time monitoring of data can realize real-time model correction,it can provide an accurate finite element model for bridge structure calculation in time.The main research contents are as follows:1.Introduction to the finite element model updating theory,general process of the finite element model is corrected;The selection of parameters in the model updating,as well as the error correction after the model made a detailed description;2.Aiming at the deficiency of the traditional method of optimal placement of acceleration sensor,this paper presents the method,combined with engineering experience to complete the Ju River Bridge acceleration sensor layout optimization;Summarize the advantages and disadvantages of different sensor performance,and introduce the layout of the whole bridge sensor;3.Aiming at the shortcomings of ordinary generalized regression neural network,an optimization method based on genetic algorithm is proposed,and apply it to the finite element model correction,detailed introduction nerve network principle and applied in the model updating process;4.Under the environment of MATLAB software,in this paper,the optimization method of generalized regression neural network is applied to the project.Through the comparison and analysis of the results,the accuracy and superiority of the neural network model updating method based on genetic algorithm is verified;5.The response surface method to correct the model,the principle and process,and apply it to rely on the project.Using MATLAB to develop a toolbox,the response surface model is modified to achieve visualization.