Local mean decomposition based on rational hermite interpolation and its application for fault diagnosis of reciprocating compressor

According to the strongly non-stationary characteristic of reciprocating compressor vibration signal, a local mean decomposition (LMD) method based on rational Hermite interpolation is proposed. Due to the excellent characteristic of shape preservation and characteristic of adjustable curve shape with parameter, Hermite rational interpolation method is used to construct envelope curves between extreme points. In order to improve the envelope approximation accuracy of local mean and envelope estimation, the local envelope curves optimization method which taken extreme value symmetry points as criterion is proposed. Algorithms and processes of rational Hermite interpolation LMD method are given. Compared with results of different interpolation methods for simulation data, superiority of this method is proved. Taken vibration signal of reciprocating compressor bearing fault as research objects, oversized bearing clearance fault is diagnosed accurately by this method, and the applicability of this method for strongly non-stationary signal is verified. ©2015 Journal of Mechanical Engineering.