Identification Of Gear Pitting Degree Based On Vibration Signal

Gear is the core basic component of mechanical equipment,and its health status directly affects the service performance and reliability of mechanical equipment.Because the gear works in the complex and harsh environment for a long time,it is easy to break down,which will lead to the failure of mechanical equipment,and even cause casualties.Pitting corrosion is one of the common failure forms of ger.It is of great significance to realize the monitoring of gear pitting corrosion.Aiming at identifying different degrees of gear pitting,the dynamic model of gear pitting was constructed,and the vibration characteristics of gear pitting were analyzed.Based on the rebuilt gear contact fatigue test bed,the identification of gear pitting of different degrees was realized based on vibration signals of the box and angular acceleration signals of the spindle.The main research work of this paper is as follows:1、Dynamic modeling of gear under pitting state.The meshing process and excitation characteristics of gear were analyzed,and the dynamic model of gear under pitting corrosion was established.First of all,combined with the meshing process of the gear,the analysis shows that the gear transmission process is periodic,which always appears alternately with single tooth meshing and double tooth meshing.The most important stiffness excitation of the gear system also changes periodically with the meshing process of the gear.Secondly,the time-varying meshing stiffness of gears under normal and pitting conditions was calculated by energy method.On this basis,the dynamic model of gear transmission system under the pitting state was established by describing the change of time-varying meshing stiffness.2、Vibration signal acquisition and feature extraction.Combined with the running condition of the gear test bench in the early stage,the gear box of the test bench was improved,and the vibration signal of the box and the angular acceleration signal of the spindle were collected at the same time,and the features were extracted and analyzed.Firstly,zero mean normalization and wavelet soft threshold denoising method are used to preprocess the original vibration signal,which reduces the interference information and improves the stability of the signal.Secondly,the time domain analysis and frequency spectrum analysis of vibration signals are carried out.The results show that the spindle angular acceleration signals are more sensitive to the change of gear pitting degree,there are less irrelevant frequencies in the spectrum diagram,and the signal characteristics change more obviously with pitting.On this basis,the five-layer wavelet packet decomposition was used to extract the energy ratio and energy entropy features of the vibration signals and spindle angular acceleration signals of the gear boxes at four different corrosion degrees respectively.Through comparative analysis,four energy nodes with high differentiation were determined as the identification feature vectors of different pitting degrees.3、Gear pitting identification based on vibration signal.Combined with the energy ratio characteristics of the wavelet packet extracted based on the vibration signal of the box and the angular acceleration signal of the spindle,the BP neural network model based on the genetic optimization algorithm was adopted to realize the classification and recognition of gear pitting of different degrees.The results show that:(1)Combined with the wavelet packet energy ratio,the BP neural network model based on genetic optimization algorithm can be used to recognize the degree of gear corrosion at different points;(2)As far as pitting corrosion is concerned,the detection method based on spindle angular acceleration signal is more sensitive,and the identification rate of gear pitting type is up to 95%,which is much higher than that of the detection method based on box vibration signal under the same working condition.Therefore,under the experimental conditions of this paper,the monitoring method based on spindle angular acceleration signal is more sensitive to the change of gear pitting state.