Updating Of Bridge Structure Finite Element Model Based On Response Surface-linear Decreasing Weight Particle Swarm Optimization Algorithm

With the continuous development of our country’s infrastructure and computer technology,the finite element model has become an important analysis to analyze engineering structure However,the finite element model contains simplifications of many factors,material parameter errors,and multiple assumptions.There is a discrepancy between the calculated results and the measured results.Therefore,it is necessary to study the improved of the finite element model and obtain a more accurate calculation model.At present,structural analysts mainly use experience to adjust the parameters in the model,which is very subjective,and the one-by-one modification of multiple parameters limits the efficiency of model updating.Response surface analysis parameters adopt experimental design,and then use explicit functions to replace the finite element model,which improves the objectivity of the selection of modified parameters and overcomes the shortcomings of repeated calling of the finite element model.Moreover,the intelligent algorithm can be combined with the response surface method to realize the simultaneous correction of multiple parameters.This article mainly conducts the following research:(1)This paper applies the particle swarm optimization to the optimization solution of the response surface objective function,analyzes the standard particle swarm optimization and proposes an improvement strategy.The particle swarm optimization with improved learning factors and the particle swarm optimization with improved weights are used to test the standard respectively.The function is solved,and the solution efficiency and accuracy of the solution are co MPared and analyzed.It is found that the linearly decreasing weight particle swarm optimization has higher accuracy and speed in the optimization solution.The key issues in the construction process of response surface method are discussed,and the model updating based on response surface method and improved particle swarm optimization is summarized.(2)This article uses the experimental design method to process the parameters,analyzes the variance based on the sample data,and analyzes the significance of the parameters combined with the F-test method to avoid the subjectivity of modifying the parameter selection.Take a simply supported single beam as an example.The central composite experimental design and F-test method were used to analyze the significance of the parameters,and the response surface model was fitted with a quadratic function without cross terms.The objective function is constructed based on the measured deflection data.The improved particle swarm optimization is used to optimize the solution,which proves that the linearly decreasing weight particle swarm optimization is efficient and stable.This quickly converges to the advantages of the global optimal solution,and improves the efficiency of the optimization solution.And the calculation error of the revised finite element model is significantly reduced,which proves the feasibility and accuracy of the model revision based on the response surface method and the improved particle swarm optimization.(3)Finally,take a continuous steel truss bridge as an engineering example.Combining its test data,the finite element model updating method based on response surface and improved particle swarm optimization are used to correct the material parameters of its upper chord,lower chord,transverse connection,diagonal web,and longitudinal and transverse beams.When the response surface is fitted,the fitting accuracy of the cross term of the quadratic polynomial function is discussed.The revised model is verified according to the test data of frequency,deflection,strain,etc.The results show that the quadratic polynomial with cross term has high fitting accuracy.The error between the calculated value and the measured value of the revised finite element model is significantly reduced,which can reflect the state of the actual structure.This correction method improves the calculation efficiency of finite element model updating,and can be used as an important means of stress analysis during bridge operation.