北京航空航天大学学报 ›› 2011, Vol. 37 ›› Issue (4): 433-438.

• 论文 • 上一篇    下一篇

测试不可靠条件下多故障诊断方法

方甲永, 肖明清, 王学奇, 禹航   

  1. 空军工程大学 自动测试系统实验室, 西安 710038
  • 收稿日期:2010-01-22 出版日期:2011-04-29 发布日期:2011-04-29
  • 作者简介:方甲永(1983-),男,河南濮阳人,博士生,fjylike@sohu.com.
  • 基金资助:

    国防十一五重点预研资助项目(51317030103)

Multiple fault diagnosis method with unreliable test

Fang Jiayong, Xiao Mingqing, Wang Xueqi, Yu Hang   

  1. ATS Lab, Air Force Engineering University, Xi’an 710038, China
  • Received:2010-01-22 Online:2011-04-29 Published:2011-04-29

摘要: 针对部队复杂系统故障诊断中存在的诊断精度低,虚警率高等问题,提出一种测试不可靠条件下多故障诊断方法.为解决系统诊断贝叶斯网络结构和概率映射表建立困难的问题,通过建立系统的多信号流图模型,从而获得系统诊断贝叶斯网络.将测试不可靠度引入概率映射表,增加了算法工程应用中的鲁棒性.利用后验概率诊断推理将问题归结为不等式约束极值问题,采用0-1规划隐数法对不等式极值问题求解,从而获得最优解.以某型导弹制导系统电子部件为例,验证了该方法对复杂系统多故障诊断的有效性.

Abstract: Aiming at low diagnosis precision and high false alarm of complex systems diagnosis, a multiple fault diagnosis method was proposed. In order to solve the difficulties of building Bayesian networks’ structure and conditional probabilities table, the system-s multi-signal flow graphs model was built. By introducing unreliability probability into conditional probabilities table, the proposed method-s robustness was enhanced. The problem was concluded as the extreme value problem with inequality constraints by using Bayes maximal posterior probability. The optimized result was obtained by the 0-1 programming implicit enumeration. The electronic equipment of guidance system was explored to illustrate the effectiveness of the proposed method.

中图分类号: 


版权所有 © 《北京航空航天大学学报》编辑部
通讯地址:北京市海淀区学院路37号 北京航空航天大学学报编辑部 邮编:100191 E-mail:jbuaa@buaa.edu.cn
本系统由北京玛格泰克科技发展有限公司设计开发