基于有向图的静止气象卫星故障检测模型

Directed graph based fault detection model of geostationary meteorological satellite

为解决基于卫星单个遥测值超限进行报警的方法会引起故障的误报和漏报这一问题,提出一种基于有向图的静止气象卫星故障检测模型。定义组合条件报警故障知识规则的描述方法和故障知识库,能够同时对卫星多个通道的遥测值进行综合分析与诊断,提高故障的诊断效率;基于有向图实施卫星故障的自动检测,省去关联故障的判别工作,减少冗余告警,提高故障检测的正确性。实际业务运行结果验证了该模型的可行性与有效性。

To solve the problem that the fault detection and alarm method by only judging a single satellite telemetry value often miss some faults and produce many false positives,a directed graph based fault detection model of geostationary meteorological satellite was presented.A description method of combined condition rules and the fault knowledge base were described,which enabled the model to analyze and diagnose satellite telemetry values on multiple channels simultaneously,and the efficiency of fault detection was increased.Automatic fault detection based on the directed graph mode was carried out,which eliminated the judging work of the associated faults,reduced the redundant alarms,and improved the efficiency of fault diagnosis.The actual business operation results verify the feasibility and effectiveness of the model.