共因失效系统可靠性的截尾次序统计量模型-JournalofNortheastern.PDF

第32 卷第8期 东 北 大 学 学 报 ( 自 然 科 学 版 ) Vol32,No.8 2011 年 8 月 Journal of Northeastern niversity( Natural Science) Aug. 2 0 1 1 1 1 2 3 武 滢, 谢里阳, 李进冬 , 李 兵 ( 1. , 110819; 2. ( ) , 110142; 3. , 313000) : - ,,, ( n ) , 3 , Monte Carlo , : ;; ;; - : TB 114.3 : A : 1005-3026(2011) 08-1165-04 Truncated Order Statistics Model for the Reliability of Common Cause Failure Systems 1 1 2 3 W U Ying , XIE Li-yang , LI Jin-dong , LI Bing ( 1. School of Mechanical Engineering Automation, Northeastern niversity, Shenyang 110819, China; 2. Shenyang Machine Tool ( Group) Design Research Institute Co. , Ltd., Shenyang 110142, China; 3. Huzhou Teachers College, Huzhou 313000, China. Corresponding author: W Ying, E-mail: wying159 @ yahoo. com. cn) Abstract: According to the common cause failure mechanism and the conception of order statistics, by means of the mapping relation between sample strength distribution in a system and order statistics, new system reliability models are proposed on the basis of load-strength interference theory and two-end truncated order statistics for serial, parallel and k-out-of-n ( F) voting systems, respectively. T he Monte Carlo simulation validates the correctness of the proposed models. Reliability calculated by the two-end truncated order statistics model is more accordant with the actual reliability of a mechanical system, compared with that obtained by traditional independent failure models and common cause failure models. Key words: common cause failure; order statistics; truncated distribution; s