HHT Analysis Of Equipment Vibration Signal And Its Application

Machinery fault diagnosis technology to guarantee the safety of equipment, reliable and efficient operation of great theoretical significance and application value. As it is easy to obtain vibration parameters through the test, so based on vibration signal processing and fault feature extraction machinery equipment is not only the most common and practical method of judging of state and fault diagnosis, but also is the most effective method.Machinery equipment's state detection and fault judged separately for the vibration signals in time domain or frequency domain processing. But vibration signal by the combination of time domain and frequency domain analysis and processing is of the latest methods of modern signal analysis, signal processing especially for the non-stationary,nonlinear signal has its unique.This paper introduces the common mechanical fault diagnosis of time-frequency analysis methods, and research a new time-frequency analysis method-Hilbert-Huang Transform (HHT), for its own algorithm to generate the end effect and false weight, etc., the paper proposes an improved method - endpoint extension and change the method of sampling rate, and combination of HHT and related analysis methods, to remove the false component, to achieve noise reduction purposes. The simulation signal and applied case studies show, the method to improve the signal to noise ratio, but also to verify the effectiveness of the method.Instance in the verification process,this article first regarding one mill reducer that have break down as the object,the fault signals collected were decomposed into a set of intrinsic mode function (IMF) using empirical mode decomposition (EMD), then the Hilbert transform is applied to each intrinsic mode function , and then the fault information is more clearly expressed by three-dimensional map and the marginal spectrum, to illustrate the method applicability in engineering applications.Then this paper applied the method of combining based on Hilbert-Huang transform and correlation analysis, through the automobile engine vibration signal measured noise processing. This method is derived by EMD decomposition of a number of intrinsic mode function (IMF) carried out the relevance of pretreatment to achieve the purpose of the signal de-noise. Then isolated the noisy modal component for Hilbert transform, the spectra obtained can be effectively extracted noise signal frequency and amplitude information.