Condition Maintenance Strategy Optimization Method For Bogie Components Of Rail Transportation Train Considering Dependencies

China rail transportation trains' maintenance strategys are mostly periodic maintenance.There are generally insufficient maintenance and excessive maintenance.In order to avoid these disadvantages,some rail transportation trains have implemented online monitoring maintenance and systematic maintenance.But in fact,on the one hand,it is difficult to obtain the monitoring information of the bogie components,which makes it is difficult to conduct online monitoring condition maintenance research for the bogie components.On the other hand,the systematic maintenance strategy ignores the dependencies between the bogie components,resulting in high cost of the systematic maintenance strategy.In view of the above problems,this paper proposes a condition maintenance strategy optimization method for bogie components of rail transportation trains considering dependencies,and implements condition maintenance based on fault data,and considers the influence of dependencies on the maintenance decision,to achieve the purpose of ensuring the reliability,availability and safety of the bogie components while taking into account economic benefits.The specific study includes the following four aspects.(1)Fault data analysis of the bogie componentsThe correlation coefficient method and the least square method are used to analyze the fault data of the axle box and the wheelset,and the Bayes theory is used to analyze the failure-free data of the axle box bearing,which fits the actual data well and provides a data support for the construction of the maintenance strategy optimization model.(2)Construction of the condition maintenance strategy optimal model for the bogie single-componentAccording to the state division method,the single-component state area is divided.We use the Markov Decision Process to make maintenance decisions,and set the minimum long-term average maintenance cost as the target,and the condition maintenance strategy optimal model for the bogie single-component is established with constraints of reliability,availability and replacement cost.The model selects the optimal maintenance mode and maintenance cycle according to the current state of the bogie single-component.It takes into account the economy,availability and reliability,and has more practical significance.It lays the foundation for the condition maintenance strategy optimization of the bogie multi-component systems.(3)Construction of the condition maintenance strategy optimal model for the bogie multi-component systemsThe optimization process of the condition maintenance strategy for the bogie multi-component systems is proposed,which leads to the idea of modular maintenance of the multi-component systems.Firstly,we use dynamic array to storage system state transition probability,and build a system condition maintenance cost decision model.Secondly,we establish a system trial maintenance strategy cost model.Then we incorporate the opportunistic maintenance state threshold,and construct the opportunistic condition maintenance cost model for the bogie multi-component systems,and modify the system state transition probability,to optimize the maintenance strategy for the bogie multi-component systems without dependencies.In addition,under the guidance of multi-component systems modularity maintenance,we consider the impact of economic dependence,structural dependence,and fault dependence on maintenance decision,and adjust the condition maintenance cost decision model for the bogie multi-component systems,to optimize the maintenance strategy for the bogie multi-component systems with dependencies.Finally,we establish the objective function of comprehensive system economy,reliability and availability for the bogie multi-component systems.The results prove that the optimization process has practical application value,and the condition maintenance strategy for the bogie multi-component systems considering dependencies is more economical and reliable,which can assist decision makers in formulating and improving the maintenance strategy.(4)Improvement of model solving algorithmWe set the judgment condition of opportunistic maintenance,and improve the policy iteration,which can be used to solve the optimization model of the opportunistic condition maintenance strategy for the bogie multi-component systems without dependencies.