Research On Analysis Method Of Gearbox Vibration Signal

The gear box is one of the most common transmission device of rotating machinery equipment,the operation efficiency of the running state of the equipment,operation effect and service life is very important.Once the gear box fault,will cause unpredictable consequences.The fault information contained rich gearbox vibration signal,but because the early gear fault signal the box is weak,easily vibration signals of external noise and other parts of the flood,therefore,accurately extract the fault information from the gearbox vibration signal,and adopt the correct method to analysis for gearbox fault diagnosis has important significance pressure.In this paper,the gear and bearing fault of the gear box is taken as the research object.The vibration signal of gearbox is analyzed by using the method of probability statistics based on the sigma criterion and the three layer decomposition method of wavelet packet:(1)The gearbox test rig was built and the vibration signal data of different speed and different working conditions were obtained through the experiment.(2)Wavelet packet denoising of the acceleration signal collected,and the amplitude of placement of probability statistics,put forward the sigma criterion based on vibration signal analysis method,research shows that if the sigma rule is small,that point to the left and right amplitude large amplitude positions are offset,the gear box does not belong to the normal state run.(3)Proposed a frequency domain analysis to determine the signal interval,then the method of wavelet packet decomposition of the interval signal,which can greatly reduce the frequency range of the signal analysis,and wavelet packet decomposition is very sensitive to signal change,the signal is decomposed into eight sub regions,observing the power spectrum,frequency,and gear box the internal parts of the meshing frequency,comparing frequency,judging by the gear box and working conditions.