Dynamic Modal Updating Based On Bolt Connection Stiffness

In practical engineering,bolted structure is widely used because of its advantages of simplicity,reliability and easy disassembly.The stiffness of bolted connection is between hinge and rigid connection,but it is usually treated as hinge or rigid connection directly in engineering.In order to improve the accuracy of numerical simulation of bolted connection structure and truly reflect the dynamic characteristics of bolted connection structure,it is necessary to determine the true stiffness of bolted connection according to the experimental data.The spring is used to simulate the half stiffness state of bolted structure.With the spring stiffness as the design variable,with the minimal residual error between the simulation data of the overall performance(natural frequency,vibration mode or displacement)of bolted structure and the corresponding test data as the objective,the optimization models are established to updating the numerical model of the bolted structure.Due to the lack of experimental data,the results of numerical model with specified bolt connection stiffness are used to replace the experimental data,and the numerical model of different bolt connection stiffness is used as the initial model to study the model updating method of bolted structure.On the platform of MSC.Patran&Nastran,the model updating program of bolted structure is developed by Python language.The main research contents are as follows:The simulation error of bolted structure mainly comes from the error of bolt connection stiffness.Through model updating,the real connection stiffness of bolted structure can be obtained,so as to improve the simulation accuracy of bolted structure.The six degrees of freedom spring are used to simulate the bolt connection to establish the simulation model of the bolted structure.With the spring connection stiffness as the design variable,with the minimal residual error of the numerical simulation results and the experimental test results as the objective or constraint,the optimization model for model updating is established.In order to improve the efficiency and accuracy of model updating,three optimization models of model updating are established.1)With spring connection stiffness as the design variable,the minimal residual error between the numerical simulation results of the eigenvector and the experimental test results as the objective,and the minimal residual error between the numerical simulation results of the eigenvalue and the experimental test results as the constraint,the objective optimization model of the vibration mode is established.2)With the spring connection stiffness as the design variable,the minimal residual error between the numerical simulation result of the eigenvalue and the experimental test result as the objective,and the minimal residual error between the numerical simulation result of the eigenvector and the experimental test result as the constraint,the natural frequency objective optimization model is established.3)With the spring connection stiffness as the design variable,and the minimal weighted value of the residual error between numerical simulation results of eigenvalues and eigenvectors and the corresponding experimental test results as the objective,an unconstrained comprehensive objective optimization model is established.Through the comparative analysis of the updating results by numerical examples,the advantages and disadvantages of the three models and the scope of application are analyzed,and a conclusion is drawn.Based on three optimization models for model updating,with MSC.Nastran software as the solver,with Python language as developing language,the model updating program of bolted structure is developed.The optimization model can be selected according to the characteristics of bolted structure.Numerical examples show the effectiveness of the optimization models for model updating and the reliability of the program.