The Wear Damage Assessment And Prediction Of Gearbox Based On Oil Monitoring And Gray Theory

Gearbox is one of the key transmission parts of engineering machinery.Breakdowns and faults are often happened because of its poor working environment,extreme load variation,and frequently periodic operations.Studies shows that the wear accounted for 80%of the gearbox parts faults.In addition,the transmission system is generally designed with non-redundancy.Once the fault occurred,it would lead to serious accidents.Therefore,in order to grasp the running state and wear trend of the gearbox from a macroscopic and accurate perspective,it is essential to study the gearbox wear damage analysis and wear prediction methods to realize the monitoring and prediction of the gear wear state and fault type.The bulldozer gearbox is selected as the research object.The damage assessment and prediction technologies of the gearbox based on the oil monitoring technology and the gray theory are studied in this paper.The main contents are as follows:(1)Experiments of the bulldozer gearbox wear damage assessment were studied.Firstly,the gearbox structure,the main wearing components,the elements of parts and the wear characteristics of particles are analyzed in depth,which lays the foundation for gearbox wear assessment through oil monitoring technology.Three kinds of oil monitoring technologies including the viscosity analysis,ferrography analysis and spectroscopy analysis were used to assess both the wear states and oil pollutions of the SD16 bulldozer gearbox.The assessment results were consistent with the actual inspection one,which proved the feasibility and accuracy of the practical application of oil monitoring technology.(2)A prediction method of bulldozers gearbox wear volume based on the modified GM(1,1)model was presented.Ferrographic analysis data of oil samples were took as the input characteristic variables,meanwhile factors,such as unequal interval of prediction data,the interference of oil change,etc.were considered in this model.Finally,a prediction method for the wearing volume of gearboxes is proposed.Research indicated.The prediction method is demonstrated reaching the "first class"precision by the posterior variance examination.The case shows that when ?=0.992,the modified model can achieve three-steps accurate prediction,while the original GM(1,1)model can only achieve one step accurate prediction.(3)The gearbox fault diagnosis test equipment was designed and built in this work,and the viscosity-temperature characteristic of 15W40 CF-4 oil was studied base on a online viscosity sensor.In order to eliminate the influence of weather,bulldozer working environment and the operator' experience on sampling data on the spot,the gearbox fault diagnosis test equipment was designed and built.The test equipment real-timely monitors the viscosity of oil by a online viscosity sensor,but this viscosity is not the standard viscosity of oil change.In practical engineering applications,the real-time oil viscosity needs to be converted into the standard viscosity when the temperature is 100? according to the oil viscosity-temperature properties.To study the variation of bulldozer lubricating oil 15W40 CF-4 viscosity-temperature characteristic,this paper measured the kinematic viscosity of the new oil,oil in mid-life and used oil in the temperature range from 40? to 100?.Regression analysis was performed to fit the experimental discrete viscosity-temperature data of the oils based on Andrade equation,Walther equation and Vogel equation.The fit goodness of these three kinds of viscosity-temperature relations was compared.The results showed that the higher accuracy is achieved when using Vogel equation.This method can promote researches on viscosity conversion of real-time online monitoring.