中国空间科学技术

›› 2011, Vol. 31 ›› Issue (4): 54-63.doi: 10.3780/j.issn.1000-758X.2011.04.008

• 技术交流 • 上一篇    下一篇

定量与定性相结合的动量轮故障可诊断性评价

刘文静1,2, 刘成瑞1,2, 王南华1,2, 王大轶1,2   

  1. （1 北京控制工程研究所，北京100190）（2空间智能控制技术重点实验室，北京 100190）
  • 收稿日期:2010-03-08 修回日期:2011-04-08 出版日期:2011-08-25 发布日期:2011-08-25
  • 作者简介:刘文静 1980年生，2009年获天津大学控制理论与控制工程专业博士学位。现为北京控制工程研究所控制科学与技术博士后。研究方向为航天器故障可诊断性评价与设计以及故障诊断。
  • 基金资助:

    国家自然科学基金（61004073, 60874055），CAST创新基金（CAST2011-6）

Quantitative and Qualitative Model Based Fault Diagnosability Evaluation of Momentum Wheel

 LIU  Wen-Jing1,2, LIU  Cheng-Rui1,2, WANG  Nan-Hua1,2, WANG  Da-Yi1,2   

  1. （1 Beijing Institute of Control Engineering， Beijing 100190) 
    （2 Science and Technology on Space Intelligent Control Laboratory， Beijing 100190）
  • Received:2010-03-08 Revised:2011-04-08 Published:2011-08-25 Online:2011-08-25

PDF (PC)

1378

被引次数

18

摘要： 为了提高动量轮的故障诊断性能，采用定量与定性相结合的方法对故障可诊断性评价方法开展研究。在分析动量轮和控制器模型的基础上，建立定量的动量轮闭环系统模型；分析动量轮的各种故障模式，将其分为测量类、驱动类和控制器类故障，并在动量轮系统模型中进行表示；依据故障到输出的传递函数是否等于0以及不同故障到输出的传递函数是否相同的思路，研究各类故障的可检测性和可分离性判断条件，对动量轮的各故障模式进行评价，给出相应的评价结果对于不可分离的故障集合，研究基于定性模型的故障可诊断性评价方法，即根据动量轮各功能模块之间的关系，结合动量轮的故障模式影响分析结果，建立故障与征兆之间的关联矩阵，通过分析各故障对应的数据，进一步给出动量轮故障可诊断性的评价结果。最后，提出故障可诊断性相关的定量指标，根据动量轮故障可诊断性评价结果，给出各故障以及动量轮的可检测度和可分离度等，为后续故障可诊断性设计提供依据。

关键词: 动量轮, 故障可诊断性评价, 故障可检测性, 故障可分离性, 卫星

Abstract: The fault diagnosability evaluation method of the momentum wheel based on the quantitative and qualitative model was proposed to enhance its fault diagnosis performance. Considering the momentum wheel model and its controller model, the momentum wheel system model was obtained and it included the influence of three types of faults, namely measurement related, driven-related and the controller-related faults. By judging whether the transfer function from the fault to the output equals to 0 or whether different faults have different transfer functions, the fault detectability and isolability discriminant conditions were obtained and applied to the momentum wheel. Some problems related to the method based on the quantitative model were concluded. To the unisolable fault sets, the qualitative model based fault diagnosability evaluation method was addressed, including the fault symptom incidence matrix establishment by considering the relationship of the function modules, the FMEA and its data analysis. Finally, the quantitative indexes were proposed as well as the diagnosable degree of the momentum wheel which offered some important information to the fault diagnosability design.

Key words: Momentum wheel, Fault diagnosability evaluation, Fault detectability, Fault isolability, Satellite