Research On Modal Parameter Identification Of Automobile Subframe Based On Underdetermined Blind Source Separation

The present study addresses the issue of complex signal testing processes and high testing costs in modal parameter identification based on positive definite and over-determined blind source separation algorithms.In order to overcome this limitation,two underdetermined blind source separation methods for structural modal parameter identification based on response signals are proposed.The research focuses on the structural modal parameter identification of a specific type of automotive rear subframe.The main research contents are as follows:(1)The blind signal separation theory and mathematical theories of four classical blind source separation algorithms are elucidated.The performance of these algorithms in separating vibration attenuation signals was analyzed.The results demonstrate that the Second Order Blind Identification(SOBI)algorithm outperforms the other three algorithms in terms of separation performance.(2)A sparse component analysis-based under-determined structural modal parameter identification method was proposed.This method exploits the clustering characteristics of structural modal response signals in the frequency-energy domain to estimate the individual modal response signal components.Additionally,the fuzzy C-means clustering algorithm is applied to estimate the mode shape matrix,and the matrix pencil algorithm was used to identify the natural frequencies and damping ratios of each individual modal response signal.The effectiveness of the proposed method was validated using numerical simulations,experimental data from a titanium alloy plate,and measured signal data from a test on a certain type of automotive rear suspension link.(3)A single-channel blind source separation-based structural modal parameter identification method was proposed,which enables the identification of structural modal parameters using a single-channel signal.The proposed method combines the Synchro Extracting Transform(SET)with the Variational Mode Decomposition(VMD)algorithm to determine the VMD algorithm parameter K.The signal was linearly reconstructed,and source signal separation was achieved by combining it with source signal separation algorithms based on signal sparsity.The effectiveness of the method was validated using simulations and experimental signal data from a titanium alloy plate.(4)A finite element simulation model of the subframe was constructed to explore modal parameter identification based on simulated signals,demonstrating the feasibility of modal parameter testing based on response signals.The modal parameter identification method was also employed for fault detection in the subframe.Furthermore,a vibration signal analysis and acquisition test platform is established to investigate the identification of free and constrained modal parameters of the subframe.The experimental results indicate that the proposed method has a maximum modal frequency identification error of less than 2% compared to the traditional frequency response function method.