Research On Reliability-centered Maintenance Decision And Support System For High-speed Train Equipments

With the development of economy in China, rail transport especially the passenger transport has come into high-speed period. High-speed trains of advanced international level have been running in large scale of China. It increases the railway passenger transport capacity of China. It also has made tremendous social and economic benefits. As more and more high-speed trains put into use, how to maintain and manage them will undoubtedly become an urgent problem which needs to be solved.The reliability-centered maintenance (RCM) idea becomes more and more widespread in the current reliability engineering field. Its maintenance ideology is that, adopting advanced detection diagnosis technology, reliability and life evaluation technology to decide the maintenance content, maintenance type, maintenance interval and maintenance level with minimal maintenance resource consumption. However, traditional RCM lacks of quantitative tools, so the maintenance decision-making process for high-speed train equipments has been made under the guidance of RCM maintenance idea. Quantitative analysis methods of equipments importance analysis, equipment life model, maintenance interval and optimization, state evaluation and prediction have been studied.High-speed train equipment maintenance strategy should be based on the equipment importance analysis and equipment failure risk analysis. In this thesis, the influence factors of equipment importance have been analyzed, and then an equipment importance analysis model based on the triangular fuzzy number (TFN) weighted logarithmic least square method has been given and corresponding evaluation process has been described. A failure risk analysis method has been given based on variable weight comprehensive analysis.The methods for the reliability model building and parameter evaluation have been studied. The combination of correlation coefficient and least-squares method has been given for the three-parameter Weibull distribution. Taking into account that, for some key equipment, their failure data is difficult to get, so evaluation method has been adopted for zero-failure data. For different models, the failure probability characteristics and evaluation methods have been studied respectively. In the process of reliability model building, imperfect maintenance mode has been introduced, and the method to get the age reduction coefficient and failure rate increase coefficient has been studied. According to the reliability model and imperfect maintenance model, different maintenance interval determination methods have been given in order to meet the different requirements for economy and reliability. Based on the analysis of optimized maintenance for single component, the multi-components maintenance optimization model has been built.The condition evaluation and prediction methods are studied deeply. At the same time through the statistical analysis of historical failure data two distribution functions are built for the failure developing process. Based on the above analysis, the optimal model for condition-based maintenance is built in consideration of imperfect maintenance and economy.A maintenance decision support system for high-speed train's equipments has been developed. The software has been programmed with VC++ as the basic tool, and the techniques such as Matlab, database, data interface have been used.