基于分层传递系统模型的航天器故障诊断方法

 针对传统的传递系统模型在航天器自主诊断系统中只能表示单一粒度诊断知识,知识表示的完备性差,导致诊断结果的分辨率不高,且诊断系统执行效率低的问题,提出了分层传递系统模型和诊断方法。为了保证不同粒度诊断知识的完备性表示,将系统模型按不同粒度和结构关系进行分层。通过自顶向下的递归搜索,先监测系统异常,再匹配故障类型,逐层缩小搜索空间,最终找出故障候选集和故障的层次关系。同时为了减小计算规模,在诊断过程中引入了分离策略,降低实时诊断计算量。应用该方法建立了某卫星测控分系统模型并进行故障仿真,结果表明该方法可以增加模型知识表示的完备性,并能有效提高诊断效率和结果的分辨率。

A Fault Diagnosis Approach for Spacecraft Based on Hierarchical Transition System Model

In order to improve the low resolution of the diagnostic results caused by the poor completeness and low diagnostic efficiency of a traditional single-granularity transition system model when modeling a spacecraft autonomous fault diagnosis system, a multi-granularity transition system model is proposed. A hierarchical strategy is used to build such a model using multi-granularity diagnostic knowledge according to the granularity and the structure. An up-to-down recursive inference algorithm is proposed to detect the failure component, confirm the fault-mode and locate the fault candidate set by reducing the search space layer by layer. Meanwhile, a separation strategy for conflict recognition is used to reduce the amount of calculation. The telemetry and command subsystem of a certain spacecraft is modeled with this approach and the result of the simulation shows that the completeness of the model, the diagnostic resolution and the diagnostic efficiency are significantly improved.