摘要
为提高航空安全,降低民用飞机运行过程中的故障风险水平,分别建立了单机风险和机队风险计算方法,指出单机风险计算的关键是确定部件故障分布及故障率,威布尔分布很好地描述了部件故障分布情况。针对部件故障机理,给出了部件故障的未检出概率、部件故障后果的条件概率及相应严重性后果,判定风险水平能否接受;机队故障风险在已知单机风险的基础上,取决于机队规模和当前飞行循环水平。基于机队运行数据预测故障飞机数量,获得不同参数分布、故障飞机数量和退役寿命下的机队风险水平。以某型发动机涡轮盘榫齿裂纹为例,通过蒙特卡罗方法,模拟单机及机队部件产生故障的时间,分析部件故障引起的单机风险水平和机队风险水平,基于运行数据给出民用飞机单机和机队在持续适航阶段的风险水平。
In order to improve aviation safety and reduce the risk level of parts failure in the process of civil aircraft opera- tion,the individual and fleet risks have been calculated. The key points of individual risk are how to confirm the failure distri- bution and rate,which turns out that Wei-bull distribution is applicable. According to the failure mechanism of part, the non-de- tection probability and the conditional probability the detect will leads to an unsafe outcome as well as related injury rate have been listed,in order to determine the risk level. Based on the individual risk,the key points of fleet risk are fleet scale and current flight circles. The number of failure flights has been predicted based on operational data. The risk value has been acquired under different operated situation such as parameter distribution ,number of failure flights as well as retirement cy- cles. The failure time of individual and fleet has been simulated by Monte Carlo method in the case of serration crack in tur- bine disc of aero-engine. The risk levels of individual and fleet about this part have been analyzed,and the two kinds risk lev- els have been given at continued airworthiness stage based on operational data.
作者
郭媛媛
孙有朝
李龙彪
GUO Yuanyuan SUN Youchao LI Longbiao(College of Civil Aviation, Nanjing University of Aeronautics and Astronautics, Nanjing 210016, China)
出处
《航空学报》
EI
CAS
CSCD
北大核心
2017年第10期150-158,共9页
Acta Aeronautica et Astronautica Sinica
基金
国家自然科学基金与中国民用航空局联合资助(U1333119)
国防基础科研计划(JCKY2013605B002)
工信部民机专项(MJ-F-2011-33)
上海民用飞机健康监控工程技术研究中心基金(GCZX-2015-05)
国防科工局技术基础科研项目(Z052013B003)~~
关键词
民用飞机
机队风险
单机风险
蒙特卡罗方法
累积分布函数
失效率
civil aircraft
fleet risk
individual risk
Monte Carlo method
cumulative distribution function
failure rate