Research Of The Fault Diagnosis For Gear Based On The Wavelet Analysis

The wavelet Analysis is a theory developed in the past decade after the Fourier analysis. It has been successfully used in fault diagnosing, telecommunication, image processing and geophysics, and so on.In order to effectively extract the feature vector of fault in the signals of gear vibration and achieve the purposes of identifying and eliminating the faults, according to the mechanism and characteristics of gear vibration, the following researches on the basis of the wavelet analysis.1) The development of technology of fault diagnosis for gear, the mechanism and characteristics of gear vibration, and the common methods of fault diagnosis for gear are discussed systematically.2) The fundamental theory of wavelet analysis is introduced on the application of fault diagnosis for gear.3) By means of the wavelet analysis, non-steady signals are analyzed, the fault feature vectors of fault are successfully extracted and this effective method is employed to identify the fault pattern.4) The low-extent noise and unexpected signal are eliminated with the wavelet threshold value method and the wavelet packet transform.5) It is presented that the expert system of the gear fault diagnosis can be designed by the use of the neural network and pattern recognition theory.In this paper, it is showed by the simulation and concrete examples represented that the wavelet analysis not only compensates the shortcomings of Flourier alternation, but also has all excellences of Fourier alternation. The fault in gear can be detected correctly and promptly, and fault pattern can be identified validly. The wavelet analysis is effective and reliable in analyzing signals; so it has its practical value and researching meaning in fault diagnosis of gear.