Modal Parameters Identification And Model Updating Of CFST Arch Bridge Based On Environmental Excitation

Because of geometry,material properties,uncertainty of boundary conditions,and the error effect caused by the simplification process during the modeling process,there is a certain deviation between the finite element model calculation results and the measured data.Therefore it is necessary to modify the initial finite element model of the structure to make the theoretical calculation results fit well with the measured results.The paper takes Yuanze Bridge in Wuhu as the engineering background.Based on the vibration response data of the bridge health monitoring system,the paper respectively uses frequency-domain peak-picking method and time-domain stochastic subspace identification method to identify the modal parameters of the bridge.Based on the design data,the initial finite element model of the bridge is established and the modified parameters are obtained through sensitivity analysis.Finally using an incomplete quadratic polynomial response surface model as an alternative model for the initial finite element model,and the optimization effect of the two opyimmization algorithms are compared.The model correction method discussed in this paper has some reference value for real bridge application.The main work and conclusions of the paper include:1.This paper introduces the health monitoring system of Yuanze Bridge in Wuhu and analyzes the experimental mode of the bridge based on the vibration response data of the bridge health monitoring.The modal parameters of the bridge are identified by the peak-picking method and stochastic subspace identification method respectively,natural fequency.The results of the two method are consistent with each other and guarantee the reliability of the results.2.According to the design data of Yuanze Bridge in Wuhu,ANSYS software is adopted to establish the initial finite element model of the bridge,analyzing the theoretical modal of the initial finite element model to obtain the calculation frequency and vibration characteritics of the bridge respectively.This provides a theoretical reference for modal analysis of vibration response data based on bridge health monitoring.3.Five modified parameters are selected by sensitivity analysis.Compared with the variance analysis method,the sensitivity method may have some disadvantages in theory,but it can reduce the number of experiment design in the response surface method and save the time cost.4.After the central composite test and parameter significance test,an incomplete quadratic polynomial response surface model is determined as an alternative model of the initial finite element model.To some extent,the response surface model can be simplified to improve the efficiency of model correction.5.The response surface method is combined with sequential quadratic programming and genetic algorithm to modify the model,and the optimization results of the two optimization algorithms are compared.Results show that the sequence quadratic programming method is superior to the genetic algorithm in the convergence speed.The optimization results of the two algorithms are equivalent.Genetic algorithm has some advantages in theory,but the concrete embodiment of its superiority remains to be studied.