基于贝叶斯的航空备件需求研究与应用  

Research and Application of Aero-Spare Based on Bayesian Method

作　　者：陈凤腾[1] 左洪福[2] 

机构地区：[1]徐州工程学院,江苏徐州221008 [2]南京航空航天大学民航学院,南京210016

出　　处：《中国民航大学学报》2011年第2期13-17,共5页Journal of Civil Aviation University of China

基　　金：国家自然科学基金项目(60672164);徐州市科技计划项目(XJ09070)

摘　　要：鉴于目前航空备件需求计算应用泊松公式存在计算结果误差较大的问题,依据历史故障数据并应用广义更新过程建立反映历史故障数据以及维修能力的可靠性分布模型,同时针对目前历史故障数据主要是小样本事件,应用贝叶斯方法并采用切片采样法对可靠性分布模型进行参数估计。最后通过实例进行分析,结果表明实际故障次数和实际故障情况较为吻合,增大了模型的应用范围。There are much more deviations in the calculation of the aero-spares demand by the Poisson formula in recently decades,so we establish the reliability distribution model to reflect the historical failure data and the maintenance capabilities with the application of the generalized renewal process.At the same time for the historical fault data for the current principal is a small sample of events,Bayesian method and slice sampling method were applied to estimate the parameter of the reliability distribution.Finally,with the example analysis,the results show that the number of actual faults and actual failures of the more consistent,increasing the application of this model.

关 键 词：贝叶斯 航空备件 故障率 切片采样法 

分 类 号：TP3[自动化与计算机技术—计算机科学与技术]