Improved EMD In The Fault Diagnosis And Verification Of Gearbox

This Gearbox as machinery and equipment in a typical connection and transmission ofpower common parts and components, mechanical transmission has been widelyused.Accordingly, the fault diagnosis of the gearbox is a very valuable research topics. In thispaper, the gear box as the research object, using improved empirical mode decomposition(Empirical Mode Decomposition referred to as EMD) method for fault diagnosis of itsvibration signal processing.Processing and analysis of vibration signals for fault diagnosis is based on a directimpact on the vibration signal noise reduction, noise reduction is good or bad to the effect ofthe fault diagnosis. This article EMD improvements is improvements in noise reduction,wavelet packet noise reduction, correlation coefficient principle and choose a differentvanishing moments db wavelet noise reduction on EMD combined with improved methods,this improvement the method enhances the noise reduction effect of the original signal.To laboratory JZQ250type gearbox fault diagnosis experiments, the improved EMDmethod is applied to the gearbox vibration signal, normal, broken teeth, bearing outer ring,inner ring failure and cage fault signalfault diagnosis, access to good diagnostic results. Inorder to prove its advantage also selected comparison method. Moreover, the paper usedwavelet packet processed spectrum and LMD spectrum with this improved EMD treatedspectrum shows a comparison of their respective advantages and disadvantages. To someextent, for future research work in the direction of improvement.Finally, the processed signaldata input established BP neural network fault identification, identification of the effect isvery good, which verifies the feasibility and accuracy of the method used in gearbox faultdiagnosis.