Parameter Identification For Linear Structures Based On Improved Sparse Signal Decomposition

Structure modeling,simulation analysis,damage identification and optimization design are based on the modal parameter identification,so the modal parameter identification is an important part of structural dynamics research,it has a good application prospect.The modal parameter identification of time-varying systems has become a new research focus along with the increasing of time varying problems in engineering structures.Therefore the method of modal parameter identification for linear systems with different excitation is studied based on sparse signal decomposition.The traditional sparse signal decomposition method exists some problems,such as the low precision of amplitude envelope and instantaneous frequency extraction.For there problems,the improved multi-scale chirplet sparse signal decomposition menthod which combined the two times fitting to extract and least square fitting instantaneous frequency is proposed.Based on the improved sparse signal decomposition method,the new method of modal parameter identification for linear system under pulse excitation is proposed,and theoretical formulas are derived for linear time-invariant and time-varying systems,at last,the three degree of freedom system simulation results show that the proposed method has high recognition accuracy and good noise immunity.For complicated environment incentives,using displacement cross-correlation function instead of impulse response function to identify the modal parameters of the linear time-invariant system combined Natural Excitation Technique(NEx T)theory,and theoretical formulas are derived for linear time-invariant systems.For the time-varying system,using the mthod which combined the improved sparse signal decomposition mthod and traditional Peak Picking(PP)mthod to identify the modal parameters of the linear time-variant system.The reaults of the simulation example and the experiment using a cantilever beam with time varying mass characteristics show that the proposed method has a high recognition accuracy and a wide range of engineering applications.