Brake Pad Life Prediction And Reliability Analysis For Electrical Multiple Unit

Brake pads are one of the most important components in the EMU braking system.In the past research,most of them focus on the formulation of materials,but few related researches are carried out from the perspective of life distribution.Find the life distribution model that conforms to the product,and calculate the reliable life of the brake pad product,which is conducive to scientific and reasonable reliability assessment of the product life,helps to formulate the EMU braking plan and replacement cycle,and improves the use efficiency and safety.while reducing the cost of use.In this dissertation,the brake pad of the EMU is taken as the research object,and the relevant domestic and foreign standards are referred to,and the failure criterion based on friction coefficient,wear amount and appearance shape conforming to reality is formulated,and the life test with environmental stress changes is designed.During the test,the failure criterion of the brake pad was corrected by the theoretical estimation value.Finally,a two-month environmental stress life test was used to obtain a total of 276 braking data of 3 groups of wear tests under each working condition.Through the half-year non-replacement fixed number censored life test,45 sets of failure life data were obtained.The traditional calculation methods of wear amount and braking distance are not very accurate.Due to the existence of eccentric wear,the determination of wear amount becomes less accurate.The analysis and summary of BP neural network and GA-BP neural network are determined to predict the wear amount and braking distance,which is more accurate and faster than the traditional calculation method.The prediction results show that the prediction error of braking distance is reduced from 7.56% to 4.13%,and the prediction error of wear amount is reduced from3.06% to 1.99%,which provides necessary preparation for correcting the error in failure life.In order to obtain reliable life,first use the non-replacement fixed truncated life failure data,draw a histogram,and further apply computer graphics of normal distribution,log-normal distribution,and Weibull distribution probability paper for modeling and comparison,and it is found that the Weibull distribution approximates The best combination effect.Three methods such as moment estimation are used to estimate the parameters of Weibull distribution,and the unbiasedness and validity are tested.The results show that the maximum likelihood estimation is the best of the three parameters.Finally,the specific life distribution of the brake pads is obtained,and the characteristic life of the brake pads in this test is obtained.The average life and reliability are 99.99% reliable life.