Rams Evaluation Of DC600V Power Inverter System

In order to meet the strong demand of transportation in our country and the needs of the development of rail transportation, improving the reliability, availability, maintainability and safety(RAMS) of railway transport equipment has become a high-profile issue. As an important technology equipment of railway transport, the RAMS evaluation of DC600V power inverter system is directly related to the safety and comfort during the train operation, but there are few studies on this problem in China.This paper is based on the development project of new DC600V inverter system which made by China Academy of Railway Sciences. Through further study of the reliability, availability, maintainability, and safety of the power inverter, it can provide theoretical foundation and basis to improve product quality, reduce the failure rate, optimize maintenance process, and improve safety level.Firstly, in this paper, the train power inverter’s main circuit and working principle were analyzed. Failure mode and effect analysis(FMEA) were used to evaluate the reliability of power inverter. The system was defined and the FMEA table was consummated, the effects of different fault parts on the system were deeply analyzed. Then the fault tree model was built up and the minmimum cut sets of the fault tree was got. At the same time, the bottom event of system failure was found out. In the end, based on the analysis, the suggestions to improve the reliability of the power inverter were proposed.Secondly, combining with fault acquisition system and the actual data in nearly five years, number of parts fault and amount of common failure types were given out. On the basis, Reliability distribution was respectively determined by using the method of traditional distribution model and neural network, and reliability parameter estimation model was established. In the meanwhile, the power inverter’s fault data distribution parameter value was obtained, failure distribution function, reliability, failure rate function and other evaluation index of the system were calculated. At the same time, the system reliability was established based on BP neural network forecasting model. Compared with the method which used the traditional distribution model, the calculation results verifyed the validity and accuracy of the adopted method. These methods for predicting the lifetime of the power inverter have a significance of guidance.Finally, the series system of power inverter by using markov process was conducted to evaluate availability and availability index was obtained. On the other hand according to actual data, maintainability index was given, such as the mean time to repair of each part, common maintenance methods are listed. Subsequently, risk assessment on the safety of power inverter was performed and high-risk links were found. Safety assessment which based on the fuzzy comprehensive evaluation model was briefly introduced and the specific solution about the system safety assessment was given.Through the study of the RAMS evaluation of DC600V power inverter, this paper can provide a reference method and basis for the evaluation of similar power electronic products or related products.