| The metro has the advantages of fast running speed,short travel time,large transportation capacity,high safety and comfort,little environmental impact and so on.With the rapid development of the metro,the requirements for its running quality are higher and higher.As one of the most critical systems of metro vehicles,the traction system must have high reliability and maintenance economy.Its running quality directly affects the stability,reliability,safety and ride comfort of metro vehicles.The traction system is a complex system.The components in the system are independent and relevant.According to the field data statistics,the traction system is one of the systems with high fault rate of metro vehicles.It is very necessary to carry out the reliability analysis and maintenance strategy research of the system,and has strong practical significance.Therefore,different methods to carry out reliability analysis and maintenance strategy research based on reliability from two aspects of system and key components is adopted.The main research contents of this paper include the following aspects.(1)A steady-state reliability analysis method of the metro traction system based on GO method is proposed.The operator in GO method is used to replace the system components.And the GO reliability model of the system is established according to the system schematic diagram.The correlation between the components in the repairable system is considered.The probability formula algorithm and the accurate algorithm of common signal in GO method are used.Then the accurate quantitative calculation and qualitative evaluation of the reliability for the metro traction system is achieved.And the weak links of the system is identified.The fault tree analysis method is used to analyze the system quantitatively and qualitatively.It proves that the GO method is reasonable and feasible.This method has clear logic,simple model structure and accurate calculation results.The logic confusion caused by the multi-level and low accuracy of the traditional fault tree analysis method is overcome.(2)The dynamic reliability analysis method of the metro traction system is proposed based on GO method and Markov process theory.The GO method is used to build the system reliability analysis model.Based on the Markov process theory,the state transition equation for the repairable two state units is established.Considering the common signal,the accurate calculation formula of the system state probability is obtained by using the state combination algorithm.And the dynamic reliability of the system and components is calculated quantitatively.A scientific basis for the maintenance strategy formulation of the metro traction system is provided by the calculation results.The deficiency of reliability analysis for urban rail vehicles traction system is solved.The dynamic reliability analysis method of the railway vehicle electrical system is enriched and developed.(3)Based on the improved fuzzy analytic hierarchy process and entropy weight method,a comprehensive evaluation method of metro pantograph fault mode criticality is proposed.The pantograph reliability models corresponding to different fault types of metro are analyzed.The entropy weight method,the improved fuzzy analytic hierarchy process and the comprehensive evaluation are combined to carry out the quantitative analysis of fault mode criticality when the service fault occur in pantograph.The subjective and objective weights of the influencing factors are determined by the improved fuzzy analytic hierarchy process and entropy weight method.Fuzzy comprehensive evaluation is used to evaluate metro pantograph fault mode criticality.The subjective and objective factors are considered in the process of determining the combination weight.The weight is more scientific.The results of comprehensive evaluation are more accurate and reasonable.The application scope of reliability assessment method is broadened.And the strong subjectivity of determining the influencing factor weight in the existing AHP is overcome.(4)A fault prediction method is proposed to predict the fault number for the power supply module of the metro vehicle traction inverter system.Through the test and simulation,the cause of the power module burning problem is analyzed.The rectification scheme is put forward.And the feasibility of the rectification scheme is verified through the test.Based on the fault number data of the power module,the idea of Grey theory,Fourier series,Markov theory and Metabolism are combined to predict the fault number.The prediction results of the original even GM(1,1)prediction model,residual first correction model,residual second correction model and the combination model of four methods are compared.The comparison results show that the combination model of four methods has higher prediction accuracy.Low precision of single prediction model based on data is solved.A scientific basis for the maintenance of traction inverter system power module and the formulation of maintenance strategy is provided.(5)The single-component preventive maintenance model of the metro traction system is proposed based on reliability.A new dynamic reliability model for components is established by statistical analysis of historical fault data and introducing the age reduction factor and fault rate increasing factor.The components are in the stage of performance degradation or random fault The influence of each preventive maintenance on the reliability of components is considered.The total cost model is established according to the maintenance modes to be adopted in the life cycle of components and various costs generated by each maintenance mode.The objective function is to minimize the unit time cost of maintenance.The optimal maintenance plan in the life cycle of components is obtained by optimizing the objective function.The proposed maintenance model is more scientific and reasonable.It not only ensures the reliability of the components,but also reduces the maintenance cost.The problem of existing maintenance strategy maintenance excess and insufficient is solved.And a method support for the group preventive maintenance strategy is provided.(6)A multi-components group preventive maintenance model of metro vehicle traction system based on reliability threshold is proposed.Based on the method of calculating preventive maintenance interval time of single component,the idea of opportunistic maintenance is introduced.Multi-components group preventive maintenance mode is adopted.It is the integration of non preventive fault maintenance,preventive maintenance,preventive renewal maintenance and opportunity maintenance.Reliability is the constraint.Restoring the components to a no new or no old state after each repair is considered.The reliability threshold of opportunity maintenance of components is a decision variable.The multicomponents group preventive maintenance model of the metro vehicle traction system is established.The model is optimized and solved.The model is efficient and economical.On the premise of ensuring the reliability of the system,the number of vehicle maintenance and the maintenance cost is reduced.A feasible and effective maintenance model and method to ensure the reliability of metro vehicles and reduce the maintenance cost is provided.Based on the reliability modeling analysis of the electrical system,reliability of the metro vehicle traction system and the key components is analyzed.The quantitative and qualitative analysis of the steady and dynamic state,the evaluation of the fault mode and criticality,and the fault analysis and fault number prediction is included.A preventive maintenance strategy model of single component and multi component for the metro vehicle traction system and its solution method from the point of view of reliability centered and maintenance cost saving are presented.An effective and beneficial analysis for the reliability and maintenance strategy of urban rail vehicle traction system is made. |
