航空学报 > 2015, Vol. 36 Issue (11): 3666-3677   doi: 10.7527/S1000-6893.2015.0078

复杂冗余系统的预测维修决策

江秀红1,2, 段富海2, 李玉峰1   

  1. 1. 沈阳航空航天大学辽宁省通用航空重点实验室, 沈阳 110136;
    2. 大连理工大学传感测控技术研究所, 大连 116024
  • 收稿日期:2014-12-24 修回日期:2015-03-17 出版日期:2015-11-15 发布日期:2015-03-25
  • 通讯作者: 江秀红,Tel.:024-89723750,E-mail:jxh_mt@163.com E-mail:jxh_mt@163.com
  • 作者简介:江秀红,女,博士研究生,讲师。主要研究方向:可靠性分析,故障诊断与健康管理。Tel:024-89723750,E-mail:jxh_mt@163.com;段富海,男,博士,教授,博士生导师。主要研究方向:精密作动器控制方法和测试计量技术。E-mail:duanfh@dlut.edu.cn;李玉峰,男,博士,教授,硕士生导师。主要研究方向:信息获取与处理。E-mail:Li_yufeng@126.com
  • 基金资助:

    航空科学基金(20130863006)

Predictive maintenance strategy for complex redundant system

JIANG Xiuhong1,2, DUAN Fuhai2, LI Yufeng1   

  1. 1. Key Laboratory of General Aviation Academy, Shenyang Aerospace University, Shenyang 110136, China;
    2. Institute of Sensor Measurement and Control Technology, Dalian University of Technology, Dalian 116024, China
  • Received:2014-12-24 Revised:2015-03-17 Online:2015-11-15 Published:2015-03-25
  • Supported by:

    Aeronautical Science Foundation of China (20130863006)

摘要:

针对多状态、多冗余复杂系统的预测维修问题,提出一种以系统可靠性为中心的预测维修设计方法。利用GO法建立了系统的可靠性分析模型,基于马尔可夫过程理论获得部件的可靠性参数随时间变化的状态转移方程,并给出了部件和系统动态可靠性的计算方法。综合部件退化度、维修成本、对系统可靠性的影响等因素提出维修优先数(MPN)的概念,以定量描述部件的维修重要性。以系统可靠性是否达标为准则设定维修时刻,以维修时刻部件的MPN为依据确定维修顺序,以维修的单位时间成本最小为目标优化维修范围和维修项目。最后以某捷联惯性导航系统(SINS)为例进行了算法验证,仿真结果表明该预测维修方法的设计是可行的、有效的。

关键词: 预测维修, 复杂冗余系统, 维修优化, GO法, 动态可靠性

Abstract:

The goal of this study is to propose a system reliability centered predictive maintenance strategy for complex structure systems with multiple redundant components and multiple states. GO methodology is applied to building the system reliability analysis model, Markov process method is used to obtain the state transition equation of components' reliability parameters changing over time, and subsequently the computing processes for dynamic reliability of components and system are presented and detailed. In order to quantify the maintenance important degree of components, a concept of maintenance priority number (MPN) is introduced here, which may comprehensively balance three importance evaluation factors, including component degradation degree, maintenance cost and the impact on system reliability. Maintenance time is determined by judging whether system reliability is up to setting threshold, maintenance sequence is determined by components MPN; meanwhile maintenance scope and the corresponding measures are optimized by an established maintenance unit time cost model. Finally, the proposed method is applied to some strap-down inertial system (SINS) and the simulation results show that the proposed predictive maintenance strategy is feasible and effective.

Key words: predictive maintenance, complex redundant system, maintenance optimization, GO methodology, dynamic reliability

中图分类号: