A mode mixing elimination method of HHT in fault detection

Abstract

As a new method of signal analysis, Hilbert-Huang Transform (HHT) has become one of the research hotspots in modern signal processing field. The main innovations of HHT are the creation of the empirical mode decomposition (EMD) method and the introduction of the concept of intrinsic mode function (IMF). The signal is decomposed as IMFs through EMD, and then we can get the instant frequency through Hilbert Transform (HT). Nowadays, the theory of HHT has been effectively applied to many practical engineering fields. However, there are lots of research to be done for the improvement and development of HHT. In this paper, the basic theory of HHT and the mode mixing phenomenon in EMD are introduced. Subsequently, two methods that can eliminate mode mixing are studied: the method of high-frequency harmonic addition and the method of ensemble empirical mode decomposition. The simulation experiments are conducted with the signal which contains interval signal and pulse of interference. Finally, a mode mixing eliminating method is proposed for the rotating blade signal. The experiment result shows that it is available for the fault detection of the rotating blade.