Abstract:For the ill-posed problem caused by noise in the identification of bolt joint’s equivalent dynamic parameters based on frequency response functions (FRFs)

an optimized identification method was put forward to obtain the accurate parameters. Nonlinear characteristics and coupling effect of the bolt joint were ignored

and the bolt joint was equivalent to spring-damper unit. Then the identification equations of the bolt joint’s tangential dynamic stiffness were established through substructure synthesis method. Through vibration experiment carried out on a specimen using LMS noise

vibration test system

the FRFs for identification were obtained. Then using the coherence functions simultaneously obtained with the FRFs to filter the FRFs to eliminate the noise’s influence on identification accuracy

and then the tangential dynamic stiffness of the bolt joint was identified based on the least square method. In the end

a harmonic response analysis was carried out on the FEM model of the bolt joint using the identified parameters. By comparing the measured FRFs with the simulated FRFs

the results show that the FEM model using the identified parameters can simulate the actual bolt joint very well and indicate that the proposed optimized identification method of bolt joint’s tangential dynamic stiffness in this paper has a good identification effect.