ASTFA-BSS方法及其在齿轮箱复合故障诊断中的应用

ASTFA-BSS Method and Its Applications in Composite Fault Diagnosis for Gearbox

自适应最稀疏时频分析(adaptive and sparsest time-frequency analysis,ASTFA)方法以分解得到的单分量个数最少为优化目标,以单分量的瞬时频率具有物理意义为约束条件,使得到的分量更加合理;结合盲源分离,提出了一种基于ASTFA的盲源分离方法并应用于齿轮箱复合故障诊断中。该方法首先利用ASTFA将单通道源信号进行分解,然后利用占优特征值法进行源数估计,根据源数重组观测信号,最后对观测信号进行盲源分离得到源信号的估计。实验结果表明,该方法可以有效地对齿轮箱复合故障信号进行分离进而实现齿轮箱的复合故障诊断。

Considering ASTFA method for decomposing the least number of single component as the optimization goal,and taking the physically meaningful instantaneous frequency of single component as constraint conditions,to make the components more reasonable,combining BSS,a BSS method was proposed based on the ASTFA,which was used to carry out composite fault diagnosis of the gearbox.In this method,the single-channel source signals were decomposed by AFTFA,then the dominant eigenvalue method was used to estimate the number of sources,and observed signals were restructured by the number of sources,finally the estimation of source signals was obtained through BSS.The experimental results show that the proposed method can separate the composite fault signal of the gearbox effectively,and achieve the composite fault diagnosis of the gearbox.