Thermal Elastohydrodynamic Lubrication Analysis And Degradation Trend Prediction Of Metro Axle Box Bearings

Rail transit construction occupies an important position in the process of urbanization in a country.The metro car axlebox bearing is a very critical foundation element in rail transit equipment.The axlebox bearings rely mainly on imports.Due to the long procurement cycle and the high cost,causing constraints on production operations.Therefore,it is urgent to carry out the localization of subway axle box bearings to improve the independent innovation capability of our enterprises and the core competitiveness of our products.In this paper,the following work is carried out for the parameter optimization,simulation analysis,performance study,running-in test,and degradation trend prediction of subway axlebox bearings.This paper introduces the theoretical basis of the study of bearings:Hertz elastic contact theory and the contact mechanics model of cylindrical roller bearings.The interaction forces between the components of the bearing are analyzed,and the static load distribution and dynamic load distribution of the bearing are analyzed.These theoretical knowledge provides theoretical support for bearing design development,performance studies,and fatigue life analysis.The CAD/CAPP/PDM integrated design system is used to optimize the product structure.The genetic algorithm is used to optimize the parameters such as the length,diameter,number of rollers and convexity of the rolling elements.The model is NNJP 5224X2-2Z.Q/P69YA subway axle box double row cylindrical roller bearing.The three-dimensional model of double-row cylindrical roller bearing was established,and its static analysis and dynamic analysis were carried out.Through the analysis of equivalent stress,overall deformation,contact force,and speed,it can provide guidance and theoretical basis for the design of bearings.Based on the research of ALE-based fluid-solid coupling bearing performance,the bearing load,contact force,friction force,temperature field,oil film pressure and oil film thickness were analyzed by the contact mechanics,lubrication performance and temperature field analysis.In particular,the changes of the central oil film and the minimum oil film thickness in the four stages of dry friction,boundary lubrication,mixed lubrication and elastohydrodynamic lubrication experienced by the subway start-up status were studied.The dynamic performance of the bearing during the start-up status is greatly affected by the slip phenomenon.By analyzing the slip rate,the dynamic performance of the bearing during the starting phase is better studied to avoid bearing failure and scratching of the raceway.The first 6 stages of the axle box bearing are modal analyzed under the condition of inner ring and no shaft constraint.The main vibration mode of the axle box bearing is the bending deformation of the cage.Using the finite element method to analyze the natural frequency,vibration mode and critical speed of the axle box bearing can provide a theoretical basis for studying the dynamic response and structural optimization of the bearing.The designed and produced double row cylindrical roller bearings were tested on a 1.2 million kilometers bench test,and then carried out on the Beijing Metro Line 8 with 1000 km of no-load and heavy-duty and 100,000 km of mainline passenger operation test.The temperature and vibration results obtained by the test are consistent with the results of the simulation analysis,indicating that the reliability of the model analysis is high,and that the bearing meets the technical conditions and current standards,and can meet the design and use requirements.The principal component analysis,phase space reconstruction and support vector machine method are combined to predict the degradation trend of the bearing.Through the PCA fusion information,the effective decay performance index is established.the Cao's method is used to determine the embedding dimension mb,the mutual information method is used to determine the time delay ?,and the particle swarm optimization algorithm is used to optimize the main parameters of the support vector machine.The predictive model of the SVM is constructed by the optimal parameters obtained to train and predict the bearing decline trend.At the same time,BP neural network,time domain index kurtosis,frequency domain index frequency root mean square,time-frequency domain IMF1 component and support vector machine are used for prediction respectively.The result comparison can be based on phase space reconstruction and support vector machine prediction results.More accurate,it provides an effective way and method for bearing life prediction.In this paper,the theoretical research is combined with the design and production process as well as the actual test process.This article is instructive for the research and development of bearings and tests.At the same time,it provides a new idea for bearing performance evaluation,which has important practical significance and engineering application value.The paper has 121 pictures,25 tables,145 references.