Research on Reliability Analysis and Application for Key Systems of Metro Vehicles
|School||Beijing Jiaotong University|
|Course||Intelligent Transportation Engineering|
|Keywords||Metro vehicles reliability fault distribution neural network|
The subway has become one of the most important mode of transport for people’s daily travel, the reliability of metro vehicles is of great significance to ensure that the Metro safe operation. This paper makes some research on the reliability technology based on the current vehicle maintenance and management situation of Guangzhou Metro Line1.First, this paper collates the failure data of Guangzhou Metro Line1Siemens train, and uses the experience-based statistical models method to determine the best failure distribution models of the door system, the signal interface system, traction/electric brake system, air-conditioning system, train control and diagnostics system, the node control circuit and auxiliary systems, and calculates their failure distribution function, reliability function and failure rate function, which has important practical significance.Then, a neural network-based system reliability analysis method was used to study the relationship of the metro vehicle systems and key subsystems. In this paper, BP neural network, RBF neural network, RBF neural network based on genetic algorithm are used to establish "the failure rate prediction model of the train key subsystem " and "the failure rate relational model of the key subsystems and the vehicle system", and the performance of each model has been compared by means of using actual data, the results show that the genetic algorithm-based RBF neural network can predict subsystem failure and describe the relationship between the key subsystems and vehicle system failure rate better.Finally, this paper developes a set of metro vehicle breakdowns and reliability analysis system, which has made "the model of failure distribution ","the failure rate prediction model of the train key subsystem " and "the failure rate relational model of the key subsystems and the vehicle system" into reality. Using the system many functions can be realized, such as vehicle breakdowns can be recorded completely, grasped detailedly and forecasted scientificly, the reliability of vehicle dynamics can be assessed, and statistics and analysis of failure become flexibility, which can help the metro staff to grasp the state and reliability of the train.