The Study Of Gear Fault Diagnosis In The Tractor's Gearbox Based On Wavelet-Genetic Algorithm

Gearbox is one of the most important units in a tractor. It is significant to study monitoring and fault diagnosis of gearbox conditions. This paper aims at the investigation of feature extraction of vibration signal and automatic diagnosis techniques of gearbox.Firstly, this paper illustrates the building of the indoor load-adding testing system applying on the study of diagnosis experiment and it's working principle.Secondly, this paper specifies the vibration mechanism of gearbox and the performances of both advantage and limitation of traditional vibration diagnosis methods of gearbox are analyzed by contrast, including the method of time domain analysis, the method of frequency domain analysis, the method of cepstral analysis and the method of the best feature parameters for frequency domain. What's more, contrasted with the method of traditional feature extraction, the method of wavelet analysis is put forward for extracting the feature parameters of the mechanical faults, including the obvious characters and the advantages of wavelet theory for feature parameters extraction in contrast with other traditional extraction methods. Furthermore, based on the wavelet analysis the feature parameter group reflecting fault feature is built.Thirdly, on the basis of feature rally extracted by wavelet analysis, this paper discusses the extraction method of the best feature parameter reflecting fault state that is derived through searching the total space of feature rally using the genetic algorithm theory and establishes the simulating blue print of tractor's gearbox representative faults so as to collect the vibration signal of different running states through the indoor load-adding testing system, including the states in gear and out of gear. Furthermore, the data derived from experiments is computed and analyzed by the theory of wavelet and genetic algorithm so that the fault pattern is recognized and by contrast with the fault patterns simulated in this experiment the diagnosis method for gearbox is certified feasible and available.