Application Research Of Hilbert-Huang Transform In The Fault Diagnosis Of Dry Vacuum Pump

With the vacuum applications continues expanding and the requirements of the vacuum applications improving, traditional oil lubrication vacuum equipments have been increasingly not able to meet stringent requirements on the market. Therefore, it has all the different types of dry vacuum pumps to meet the need of the oil-free and cleaner vacuum. It's very necessary to research the fault diagnosis on the dry vacuum pump.Traditional fault feature extraction methods based on the assumption of linear and station coundn't accurately time-frequency various feature of non-stationary signals.So,it presents Hilbert-Huang transform (HHT) to extract fault feature of non-stationary signals for mechanical equipment,the method is a new method to deal with nonlinear and non-stationary signals.It states the development present situation and significance of Revolving Machine's mechanical malfunction diagnose system; It particularly introduces the background and the development present situation of the HHT.It is expatiated the basic segment of the fault diagnosis and some kinds of normal fault diagnosis of rotary machinery. It introduces common diagnosis method of rotary machinery fault and its advantages and disadvantages. Several commonly used time-frequency methods are briefly introduced.The system introduces the basic theories of the HHT. It expatiated that the significance of the Empirical Mode Decomposition(EMD).It analyzes the Intrinsic Mode Function(IMF) which can make the instantaneous frequency meaningful from the definition of the instantaneous frequency. It analyzes the completeness and the local orthogonality of EMD.And it sums up the main problems of the HHT in currently.We used a signal interpolation method to reconstruct the signal on sampling lowly against the impact of EMD algorithm.The filter arithmetic fetches in the part filter arithmetic.It researched the insufficiency of the Nurel Networks extension method and the Mirror Periodic extension method.The author introduces the Variable Length Mirror Envelope Extremum extension method on disadvantages.The Matlab programming language is applicated in the HHT algorithm and traditionary Hilbert transform envelope demodulation method.The Hilbert-Huang transform is applied to the fault diagnosis of the dry vacuum pump, from the test analysis,compared with the method of the Hilbert transform envelope demodulation.The results showed that HHT in fault characteristic extraction has higher resolution and accuracy.