Fault diagnosis of natural gas compressor based on EEMD and Hilbert marginal spectrum

Abstract

The paper utilizes ensemble empirical mode decomposition (EEMD) and Hilbert marginal spectrum for the fault diagnosis of the reciprocating compressor on the offshore platform of WZ12-1, aiming at the non-stationary and nonlinear characteristics of vibration signals collected from the faulty compressor. First, the EEMD algorithm self-adaptively anti-aliasing decomposes the vibration signal into a set of intrinsic mode function of different frequency bands. Then, the Hilbert marginal spectrum with some advantages in frequency resolution is used to extract the fault feature. Next, the proposed method succeeds in diagnosing the fault of the reciprocating compressor. The results show that the proposed method is feasible.