The Acoustics Fault Diagnosis Of Gear Box Based On EMD

The whole production process cannot work normally if the mechanical equipmentbreakdown. It also causes the great economic losses and even endangers the personal safety.Mechanical vibration must be accompanied by sound, sound signals and vibration signalswith good correlation. For example, the energy distribution of the collected vibration andsound signals are bound to change if there is something wrong with the mechanicalequipment. The characteristics of fault information can be extracted through analyzing thevibration or sound signals. The signal acquisition of acoustic fault diagnosis technology hasthe advantages of ease and non-contact measurement. The technology is wildly used becauseit doesn’t affect the normal work, and it can be used in the occasions of difficult to measurethe vibration signal. However, the technology also has the disadvantages of extracting featureinformation difficultly in that the sound signal is very low signal to SNR of non-stationarysignals. Therefore, proposition of this paper based on the EMD acoustic fault diagnosis. Themain contents of this paper are as follows:（1）Study on the vibration mechanism of gears and bearings in the gear transmissionsystem, the common type of fault, the reason and the characteristic.（2）The relationship between the sound signals and vibration signals are analyzedtheoretically. The collected sound signals and vibration signals are proportional relation whenthe acoustic sensor arranged in a specific position. According to this feature, the layout ofmultiple measuring points are analyzed and compared, and finally elected the best measuringpoint which laid the foundation to extract the sound signals processing and the characteristicsof the gearbox fault.（3）Study on the method and characteristic of EMD and its simulation analysis verifythe separation effect of EMD on the simulation of multicomponent signal is very good.（4）Study on the application of EMD frequency family separation method in gear fault feature extraction to obtain better diagnostic by used EMD method to extract gear faultfeature signal from the audio signal.（5）Study on local Hilbert marginal spectrum and its application in extraction of rollingbearing fault feature to obtain better diagnosis effect by used wavelet and EMD methodextracted fault signal of bearing inner ring and outer ring from the sound signal.