中国安全科学学报 ›› 2021, Vol. 31 ›› Issue (11): 54-63.doi: 10.16265/j.cnki.issn 1003-3033.2021.11.008

• 安全工程技术 • 上一篇    下一篇

适航指令设计制造缺陷案例关联分析*

曹惠玲 教授, 胡彦杰   

  1. 中国民航大学 航空工程学院,天津 300300
  • 收稿日期:2021-08-11 修回日期:2021-10-04 出版日期:2021-11-28 发布日期:2022-05-28
  • 作者简介:曹惠玲 (1962—),女,山西临汾人,博士,教授,主要从事航空发动机性能分析、故障诊断等方面的研究。E-mail:hlcao@cauc.edu.cn。
  • 基金资助:
    中央高校基本科研业务费项目(2020YJS003)。

Association analysis on CAD design and manufacturing defect cases

CAO Huiling, HU Yanjie   

  1. School of Aeronautical Engineering, Civil Aviation University of China, Tianjin 300300, China
  • Received:2021-08-11 Revised:2021-10-04 Online:2021-11-28 Published:2022-05-28

摘要: 为有效预防航空安全事件的发生,找到事件发生的潜在关联,分析中国适航指令(CAD)所反应的设计制造缺陷案例的关联性。首先,对CAD制定属性提取规则;然后,采用不完备信息系统的粗糙集模型,将故障发生的危险等级作为决策属性,并约简条件属性;最后,通过FP-growth算法挖掘易损部位、事件原因、事故等级、处理措施等规则,找出引起不适航或超出运行条款导致不能安全运行因素之间的内在联系。研究表明:经过粗糙集的约简,案例关联性分析能够提取到置信度高的规则,有助于适航管理机构进行相应决策。

关键词: 适航指令(CAD), 设计制造缺陷, 粗糙集, FP-growth算法, 不完备信息系统, 关联规则

Abstract: In order to effectively prevent aviation safety incidents and find potential correlation between them, correlation of design and manufacturing defect cases reflected by CAD was analyzed. Firstly, rules of attribute extraction were developed for CAD. Then, risk levels of failure was taken as decision attributes to reduce condition attribute by using rough set model of incomplete information system. Finally, FP-growth algorithm was used to discover rules of vulnerable parts, event causes, accident levels, and treatment measures, so as to find out the internal relations among factors that cause unairworthiness or unsafe operation due to incompliance with operation regulation. The results show that association analysis of cases can extract rules with high confidence after rough set reduction, which is helpful for airworthiness management agencies to make corresponding decisions.

Key words: China airworthiness directive (CAD), design and manufacturing defect, rough set, FP-growth algorithm, incomplete information system, association rules

中图分类号: