Research On The Maintenance Strategy And Fault Diagnosis Of The On-board Equipments In Metro Train Based On RAMS

Metro train is one of the best public traffic modes, which is convenient, stable, fast, low noise and clean. In order to solve urban traffic jams, urban metro has become an important trend. In recent years, the development of China’s urban metro has entered a rapid development period. There are more than30cities planning to build metros and the investment on metro will be more than1trillion, which will improve the city transportation greatly. Metro train, the core part of the metro transportation, is the carrier of passengers, its reliability, availability are must be improved to reduce the maintenance costs and risks. Currently, the metro train operation management exist the following problems:the maintenance strategy for the metro train is formulated based on experiences; the fault diagnosis of metro trains mostly relies on the subjective judgment of the maintenance staff, diagnostic systems rely on foreign technology, as well as, the management platform of data and maintenance program are lacked.This thesis is carried out based on the project:development of Guangzhou Metro fault diagnosis and expert system. According to the metro train system architecture, equipment reliability, availability, maintainability and safety assessment theory are researched, the relevant mathematical models are established and improved. Further more, the problem of improper maintenance and fault location are solved, a fault diagnosis system and expert system are developed which have been put into practical application. The main research contents include:First of all, the common failures of metro trains and the failure components of equipment are counted and classified. Moreover, using the improved AHP (Analytic Hierarchy Process) and Monte Carlo method analysis of related data, establish a quantitative evaluation system of components, provide the basis for determine the component maintenance way.Next, this thesis analyzes the characteristics of the component running data, and further proposes the modeling method of the component reliability and safety. These methods provide a theoretical basis for establishing average reliability model of the system and formulating the maintenance strategies. Method was applied to fault data analysis and processing, some components’ reliability models are obtained. With the established model, the reliability and safety of the equipment can be evaluated accurately. Following, the influences on the reliability, availability, safety and maintenance costs, generated by the maintenance interval and repair modes, are analyzed. And, the multi-parameter dynamic model of the RAMS is proposed in the thesis. More important, the thesis proposes an improved chaotic adaptive evolutionary algorithm, by which, the multi-parameter dynamic model can be solved and the maintenance strategy can be obtained. Compared the solution set which obtained by the improved algorithm, Multi-Objective Particle Swarm Optimization (MOPSO) algorithm and Non-Dominated Sorting Genetic Algorithm (NSGA2), the improved algorithm obtained solution set is better. This thesis made reliability and maintenance cost as the optimization goal, availability and security as constraint conditions, optimize the maintenance strategy. The optimal solution set provide reference for reasonable maintenance strategy.Finally, On the basis of analyzing the problems existing in diagnostic system, the thesis designs and develops a set of fault diagnosis system and expert system for the metro train. Fault diagnosis system is used to record what information and vehicle status; in the expert system, the fault database, case base, knowledge base and model base are established. Database standardizes the RAMS data recording way, case base is to help establish case model, accumulate experience for fault diagnosis. Combined with the "emergency brake fault" case, the case retrieval and matching process are analyzed. The system realizes the fault location, and gives maintenance recommendations, effectively improve the maintenance efficiency. Fault diagnosis and expert system designed in this paper have advantages in low cost, good performance, high reliability and so on, and have been passed through the relevant national and European industry standard type tests, and used in practice. The fault diagnosis and expert systems fill the domestic blank at related equipment.