中国空间科学技术 ›› 2020, Vol. 40 ›› Issue (6): 56-67.doi: 10.16708/j.cnki.1000-758X.2020.0072

• 研究探讨 • 上一篇    下一篇

基于动态故障树的卫星可靠性分析

王许煜,胡敏,张学阳,赵玉龙,李玖阳   

  1. 航天工程大学,北京101400
  • 出版日期:2020-12-25 发布日期:2020-11-25
  • 基金资助:
    国家自然科学基金(61403416)

Satellite reliability analysis based on dynamic fault tree

WANG Xuyu,HU Min,ZHANG Xueyang,ZHAO Yulong,LI Jiuyang   

  1. Space Engineering University,Beijing101400,China
  • Published:2020-12-25 Online:2020-11-25

摘要: 针对卫星系统的可靠性问题,提出了一种基于动态故障树的卫星可靠性分析方法。采用马尔可夫链和二元决策图相结合的分析方法,建立卫星的电源、姿轨控和推进3个分系统的动态故障树模型,在此基础上得到卫星的随机故障模型,并综合考虑损耗故障建立卫星可靠性模型。利用蒙特卡洛仿真对随机故障模型进行评估分析,并将其与Weibull分布模型进行性能比较,仿真结果表明该方法能够有效分析卫星的随机故障,具有计算精度高、效率高的优点,并能更有效地模拟卫星部件随机故障的动态行为。

关键词: 卫星可靠性, 随机故障模型, 马尔可夫链, 二元决策图, 蒙特卡洛仿真, Weibull分布

Abstract: Aiming at the reliability problem of satellite system, a method for satellite reliability analysis based on dynamic fault tree was proposed. The dynamic fault tree model of three subsystems of satellite power, attitude and orbit control and propulsion was established by combining Markov chain and binary decision diagram. On this basis, the random fault model of satellite was obtained, and the reliability model of satellite was established by taking the depletion fault into consideration. Monte Carlo simulation was used to evaluate and analyze the random fault model, and its performance was compared with Weibull distribution model. The simulation results show that this method can effectively analyze the random fault of satellite, has the advantages of high computational accuracy and high efficiency, and can simulate the dynamic behavior of random fault of satellite components more effectively.

Key words: satellite reliability, random fault model, Markov chains, binary decision diagram, Monte Carlo simulation, Weibull distribution