The Analysis Of Fault And Research On Maintenance Strategy Based On Locomotive System

The safety,reliability and maintainability of high-speed trains have become an issue of growing concern.The trains reasonable repair and maintenance strategy,which is not only related to the running safety,but also related to the running benefits,is the foundation of the sustainable development of high-speed railway.The reliability-centric maintenance strategy is a maintenance method that is commonly used in the world.This is a method to determine the reasonable maintenance strategy and cycle which based on field statistics of fault data information and analyzes of the failure mode and life distribution of each system.Based on a long time of fault information recorded on the vehicle loading system of the HXD1C locomotive,this paper analyzes the failure mode and life distribution of each system and analyzes their maintenance strategies according to the different life distributions.The specific work results are:First of all,I sorted out the fault data,divided into frequent fault data and less fault data,and get the life data.For frequent fault data,I used the method of parameter estimation and hypothesis testing.Respectively,based on exponential distribution,logarithmic normal distribution and Weibull distribution to fit life distribution,and according to K-S statistics to determine the best distribution type.Based on this,the motor,brake system as examples were analyzed,the fault types can be divided into three categories.For less fault data,I used the way of matching distribution curve method.The weighted least squares estimation is performed based on these three distributions,and the optimal life distribution type is determined according to the value of goodness of fit.Then,I studied how the failure of components affect the system failure from two aspects.The first is to conduct hazard analysis,taking the motor and the transformer as an example,analyzed the degree of harm in each failure mode.The second is to study the law of failure,taking the motor as an example,analyzed the influence of the failure of the six parts on the motor system and obtained their relationship expressions,that is,the failure regularity of the system is affected by the type of failure distribution of the component,and the influence factor is related to the fault frequency.Finally,based on the different fault types,the three maintenance strategies of the principle of reliability model,economic principle model and reliability promotion principle model are studied respectively,and the best maintenance methods and maintenance periods under different strategies are obtained respectively.Then according to the actual situation and requirements,we can flexibility choose the best maintenance methods.The innovation part of the paper is embodied in:First,when using the distribution curve method to estimate the failure probability,the product limit estimation method is proposed to estimate the failure probability for the case of very few failures,unlike the conventional way,used the Bayesian estimation method for the case of no-failure data and single-failure data.Second,when studying how components fail affect system fail,based on the ideas of the failure of the components led to the failure of the system and the system fail in a variety of modes and what kind of relationship between component failure mode and system failure mode,I estimate the overall failure distribution of the system in four ways,and get the approximate expression of their relationship.Third,I combined the two aspects of safety and effectiveness,and develop a maintenance method based on the principle of reliability improvement,and measured the significance of maintenance costs by the proportion of increasing of maintenance costs and the proportion of reliability to determine the maintenance cycle.