Research On Life Prediction Of Shearer Spindle Bearings

At present,shearer is mainly used to mine coal bothhere and abroad,the spindle bearing is an important component of the shearer,and its fatigue life directly affects the working efficiency of shearer.Therefore,the research on the fatigue life of the shearer spindle bearing has important practical significance and theoretical value.In the past study of bearing fatigue life,most scholars predict the bearing life mainly through the stress value obtained from bearing dynamics simulation,however,as one of the important factors affecting the life of the spindle bearing of the shearer,the working condition parameters of the bearing are often ignored in predicting the bearing life,therefore,this paper mainly studies the influence of different working condition parameters of the bearing on the bearing stress in order to predict the bearing fatigue life more efficiently,it is applied to the bearing life prediction.In this article,NJ2232 bearing of shearer spindle bearing is taken as the research object,the main research work is as follows:(1)Establish the sample set of bearing stress value under different working conditions.Firstly,the force analysis of the shearer spindle bearing is carried out,the theoretical value of the bearing contact stress is calculated according to Hertz's theoretical formula,the three-dimensional model of the bearing is established,and the stress value of the bearing under different working conditions is obtained through the finite element analysis;(2)An improved BP neural network data mining model is established to increase impulse term,self-adaptive learning rate and multi-layer training method to determine the network structure,and the importance order of the different working condition parameters of the shearer spindle bearing affecting the bearing contact stress and fatigue life is analyzed,which provides a theoretical basis for effectively improving the fatigue life of the bearing;(3)According to the fatigue damage accumulation theory and fatigue life calculation,the bearing fatigue life is predicted,considering the influence of various working conditions of the bearing on the stress value,the analysis results are more consistent with the actual operation of the bearing,thus improving the accuracy of bearing life prediction.The method studied in this article can provide technical support for the selection and operation of the working condition parameters of the shearer spindle bearings.According to the stress cloud diagram obtained by finite element analysis,it can beseen that the bearing components are subjected to different contact stresses,which is consistent with the actual stress situation of the bearing.The data mining analysis shows that the biggest influence factor of contact stress is load,and its influence factor is 0.63,relatively speaking,speed and lubrication have little influence on it,and the influence factors are 0.25 and 0.12 respectively.The improved BP neural network algorithm analyzes the impact of bearing operating parametersis more suitable for the actual operating conditions of the bearing,it is more accurate than the traditional control variable method to study the bearing working condition parameters.Applying the nominal stress method,according to the S-N curve of the main bearing and miner's linear fatigue damage accumulation theory,the fatigue life of the bearing under different working conditions is predicted to be 2.42 years(24-hour working system).The fatigue life of the bearing is predicted to be 2.15 years(24-hour working system)on the basis of fully considering the influence of working conditions on the contact stress.