Misalignment diagnosis of rotating machinery through vibration analysis via the hybrid EEMD and EMD approach

The objective of this research is to investigate the feasibility of utilizing the hybrid method of ensemble empirical mode decomposition (EEMD) and pure empirical mode decomposition (EMD) to efficiently decompose the complicated vibration signals of rotating machinery into a finite number of intrinsic mode functions (IMFs), so that the fault characteristics of the misaligned shaft can be examined in the time-frequency Hilbert spectrum as well as the marginal Hilbert spectrum. The intrawave frequency modulation (FM) phenomenon, which indicates the nonlinear vibration behavior of a misaligned shaft, can be observed in the time-frequency Hilbert spectrum through the Hilbert-Huang transform (HHT) technique. The fault characteristic of shaft misalignment is also featured in terms of the amplitude modulation (AM) phenomenon in the information-containing IMF components that are extracted by the significance test. Through performing the envelope analysis on the information-containing IMF, the marginal Hilbert spectrum of the envelope signal of this IMF component exhibits that the level of shaft misalignment is presented by the level of AM in the IMF. A test bed of a rotor-bearing system is performed experimentally to illustrate both the parallel and angular shaft misalignment conditions as well as the healthy condition. The analysis results show that the proposed approach is capable of diagnosing the misaligned fault of the shaft in rotating machinery and providing a more meaningful physical insight compared with the conventional methods.