论文部分内容阅读
结合低轨通信星座系统自主运行发展的需求,提出一种适合于星上自主处理的分布式网络故障诊断算法(DFDS).DFDS算法基于系统级故障诊断理论(SLD),给出M-概率的分布式测试模型,依据小概率忽略的诊断规则,通过测试图预修正、二次判决后处理等风险消减措施,达到识别故障的目的.仿真实验和分析表明该算法能够有效地提高故障诊断的正确率和完全率,同时在诊断开销和诊断时延方面也较同类算法性能有明显优势.
Combining with the demand of autonomous operation of low-Earth communication constellation system, this paper proposes a distributed network fault diagnosis algorithm (DFDS) suitable for on-board autonomous processing.The DFDS algorithm is based on the system-level fault diagnosis theory (SLD) The distributed test model achieves the goal of identifying faults by means of risk reduction measures such as pre-test of test chart, post-process of second judgment, etc. According to the diagnostic rules ignored by small probabilities, the fault diagnosis is achieved by simulation test and analysis.The results show that the algorithm can effectively improve the fault diagnosis Rate and completeness rate, at the same time in terms of diagnostic overhead and diagnostic delay than the performance of similar algorithms have obvious advantages.