关键词: 工作模态参数识别, 稀疏成分分析, 欠定, 模态振型评估, 独立成分分析

Abstract: Aiming at the problem that the modal parameter identification method based on independent component analysis can identify fewer modal numbers,a method based on sparse component analysis was proposed.The basic idea of the method was to find the modal mode response between the mode shape and the mixing matrix from the modal mode of linear time-invariant small damping structure,the properties of the modal coordinates and the basic assumptions of sparse component analysis.A one-to-one correspondence with sparse components and the problem of underdetermined working modal parameter identification were transformed into a sparse component analysis problem.From the problems existing in the identification of working modal parameters based on traditional blind source separation and the process of sparse component analysis,the framework and process steps of working modal parameter identification based on sparse component analysis were established.For the problem of interpretation and evaluation of the identification mode of underdetermined working modal parameters,the characteristics and evaluation methods of identifying vibration modes under underdetermined conditions were proposed.The adaptation range of identifiable modal number,modal missing and false modal and method was discussed and compared with the independent component analysis method.The identification of working modal parameters under the simulation data set of 5 degrees of freedom showed that the modal parameter identification method based on sparse component could identify the modal parameters exceeding the number of vibration response sensors,and the number of modals identified by the independent component analysis method.When the number of recognition modes was less than or equal to the number of degrees of freedom,the recognition result was a true mode,and a false mode existed in the recognition result if the number of recognition modes was greater than the number of degrees of freedom.

Key words: operational modal parameters identification, sparse component analysis, underdetermination, modal shape estimation, independent component analysis