Research On Modal Parameter Identification Of Automobile Suspension Parts Based On Blind Source Separation Algorithm

Modal analysis is an important means to analyse the NVH performance of automobile.As an important supporting component,suspension parts ensure the safety and comfort of vehicles.How to solve modal parameters is the key point of modal analysis for suspension parts.Traditional modal parameter identification technology is usually based on known input and output of the system to obtain modal parameters.In some specific scenarios,such as the vehicle operating conditions,the input parameters are difficult to measure,and the traditional modal recognition method is not easy to solve such problems.In order to identify modal parameters only according to the output response,this paper focuses on the theoretical and experimental research on identifying modal parameters based on blind source separation algorithm.Based on the analysis of the basic theory of blind source separation,this paper studies three classical algorithms,including Fast ICA algorithm,JADE algorithm and SOBI algorithm,which are suitable for vibration signal separation.The separation of the three vibration related mixed signals is studied by using these three classical algorithms.Except that the order and amplitude of the separated signals are different from the source signals,the waveforms of the separated signals and the source signals are the same.The model of blind source separation algorithm and vibration response model are analyzed.The modal expansion and blind source separation algorithm are compared.The blind separation algorithm can be applied to the modal parameter identification by theoretical analysis.The discrete model of three degree of freedom vibration system is established,and the three classical blind source separation algorithms are used to separate the displacement output signal of three degree of freedom system and identify its modal parameters when considering damping,noise,excitation form and other factors.The robustness and accuracy of the three BSS algorithms for solving modal parameters are investigated under different conditions(different damping,white noise signals with different signal-to-noise ratios and random vibration excitation).The platform of modal parameter identification based on blind source separation algorithm is built,and the modal parameter identification of the upper link of suspension is studied under free vibration and random vibration.The accuracy and reliability of the upper arm modal parameter identification results based on the blind source separation algorithm are verified by taking the modal parameter value of the traditional modal test as the reference standard.