Fatigue Life Reliability Research Of High-speed Railway Bearing Based On Multi-source Information Fusion

Due to the continuous reform and progress of China s high-speed railway technology in recent years,higher requirements have been placed on the performance of each component of high-speed train.As an important mechanical component of high-speed train transmission system,the reliability of axle box bearing has a great influence on the transmission performance of high-speed train.The axle box bearing has complex structure and bears complex load impact in operation.The key technology to improve its reliability is to establish a high-precision fatigue life evaluation and prediction model for axle box bearing.This paper takes the high-speed railway axle box bearing--double row tapered roller bearing as the research object.Considering the uncertain factors of manufacturing and operating conditions to study its fatigue life and reliability.The specific work is:(1)The fatigue life reliability assessment of high-speed railway axle box bearings is based on finite element simulation and failure physics accelerated life model.Firstly,analyzing the failure mode and load of the axle box bearing,and establishing a random finite element simulation model to analyze the fatigue life distribution of the axle box bearing under the influence of random loads and manufacturing errors.On this basis,considering the influence of random loads on accelerated life factors,a physical accelerated life model for failure under random loads is established.The cumulative distribution of fatigue damage of axle box bearings is further calculated according to the linear accumulation law of fatigue damage,and the fatigue reliability of axle box bearings is evaluated.(2)The fatigue life reliability assessment of high-speed railway axle box bearings is based on statistical accelerated life models.Firstly,a statistical accelerated life model with time-varying ratio of the distribution function of the axle box bearing hazard rate under different stress levels is established.Secondly,according to the life data of the bearing,evaluating the proposed statistical accelerated life model parameters of the axle box bearing,and calculating the fatigue life distribution of the axle box bearing under different working conditions.On this basis,the fatigue reliability of axle box bearings is evaluated.(3)The fatigue life reliability assessment of high-speed railway axle box bearings is based on vibration signals.Firstly,the collected bearing vibration signal information is de-noised,feature extraction and feature selection.Secondly,a prediction model of cyclic neural network is established to predict the bearing life.The established cyclic neural network life prediction model is further used to calculate the fatigue life and distribution of axle box bearings under different working conditions,and on this basis,the fatigue reliability of axle box bearings is evaluated.(4)The fatigue life reliability assessment of high-speed railway axle box bearings is based on D-S evidence theory fusion data.Failure physical data,statistical life data and vibration signals reflect the operating state and life of the bearing from different angles.The D-S evidence theory is used to fully excavate data and information,and these three types of data are combined to evaluate the fatigue life reliability of axle box bearings under different working conditions,and to compare and analyze the evaluation results of various methods.