北京航空航天大学学报 ›› 2014, Vol. 40 ›› Issue (5): 602-607.doi: 10.13700/j.bh.1001-5965.2013.0361

• 论文 • 上一篇    下一篇

基于状态条件概率的设备剩余寿命预测

张继军, 邓力, 马登武, 曹文静   

  1. 海军航空工程学院, 烟台 264001
  • 收稿日期:2013-06-25 出版日期:2014-05-20 发布日期:2014-06-04
  • 作者简介:张继军(1979- ),男,山东安丘人,工程师,zjj2011pp@126.com.
  • 基金资助:

    国家自然科学基金资助项目(61203168);总装备部预研基金资助项目(9140A27020212JB14311)

Remaining useful life prediction for equipment based on conditional probability of states

Zhang Jijun, Deng Li, Ma Dengwu, Cao Wenjing   

  1. Naval Aeronautical and Astronautical University, Yantai 264001, China
  • Received:2013-06-25 Online:2014-05-20 Published:2014-06-04

摘要: 针对机载设备剩余使用寿命预测中存在的不确定性因素,建立了基于状态条件概率分布的机载设备剩余寿命模型.首先,引入状态条件概率矢量对隐马尔科夫模型(HMM,Hidden Markov Model)进行不确定性改进,并推导了其计算形式.其次,给出了近似确定状态退化转移时间的方法,由此得到了以状态条件概率矢量为协变量的条件可靠度函数及剩余寿命模型.最后,以飞机发动机温控放大器为应用对象进行仿真计算.仿真结果表明该模型预测精度高,能够较大程度地降低不确定性因素的影响.

Abstract: To conquer the question of uncertain factors in the remaining useful life (RUL) prediction of airborne equipment, a RUL model was presented based on the conditional probability distribution of states. Firstly, the conditional probability vector of states was introduced to improve the ability of hidden Markov model (HMM) to deal with uncertainties, and its computation form was deduced. Secondly, a method was presented to approximately determine the degradation transfer time of states, thus the conditional reliability function and RUL model were obtained, in which the conditional probability vector of states was taken as the covariate. Lastly, the aero-engine temperature control amplifier was applied to validate the model. The results show that the prediction precision of this model is high, and this model can reduce the influence of uncertain factors greatly.

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