基于ACMD和相空间重构的转子不平衡故障诊断

Rotor Unbalanced Fault Diagnosis Based on ACMD and PSR

针对转子系统发生频率较高的不平衡故障,提出了一种基于自适应调频模式分解(Adaptive Chirp Mode Decomposition,ACMD)和相空间重构技术相结合的故障诊断方法。首先,通过ACMD将不平衡故障信号分解成若干个分量,使用相关系数进行模式选取;其次,对筛选出的最相关模式进行三维相空间重构(3-dimensional phase space reconstruction,3D-PSR)特征的提取;然后,基于修正后的k-NN分类方法(基于最小欧式距离)原则实现对转子系统不平衡故障的诊断。最后,在转子系统故障试验台上进行实验验证,和已有方法相比,提出的方法可以更有效提取出不平衡故障特征,并且能在样本较小情况下具有更高的准确率。

In view of the high-frequency unbalanced faults of the rotor system,a method based on adaptive chirp mode decomposition(ACMD)and phase space reconstruction(PSR)technology is proposed.Firstly,the unbalanced fault signal was decomposed into several components by ACMD,and the correlation coefficient was used to select the mode.Secondly,the features of 3-dimensional phase space reconstruction(3D-PSR)were extracted from the most relevant modes selected.Then,based on the modified K-NN classification method(based on minimum Euclidean distance)principle,the unbalanced fault diagnosis of the contra subsystem was realized.Finally,the experimental verification was carried out on the rotor system fault test bench.Compared with the existing methods,the proposed method not only can extract unbalanced fault features more effectively,but it also has higher accuracy in the case of small samples.