摘要
空间站实验柜密封箱体在通过抽真空、充氮气进行排废气时一旦失控,将产生箱体失压或过压的风险,对排废气过程进行安全监控和故障识别,将面临多参数耦合以及时序逻辑等复杂事件处理问题。将大数据流处理技术Flink及其复杂事件处理(CEP)库应用在排废气安全监测上,设计出相应异常模式序列对排废气过程监测的数据流进行匹配,从数据流中实时检测出复杂事件序列。在仿真测试条件下,基于Flink CEP的监测程序故障检测率和隔离率均达100%,能有效剔除传感器瞬态异常导致的虚警。
Once the vacuuming and nitrogen filling during exhaust gas discharge(EGD)for the sealed box of experimental rack onboard space station get out of control,the sealed box might encounter risks of pressure loss or overpressure.The streaming processing technology Flink and its Complex Event Processing(CEP)library were implemented for the EGD safety monitoring.First,the corresponding abnormal pattern sequencies were designed to match with the streaming collected during EGD process.Thus,the complex events were detected in real time from the streaming by the program developed through Flink CEP.Testing showed that Flink CEP could detect and isolate all the simulated faults,and reject the false alarms caused by transient anomalies of sensors.
作者
施建明
王伟
王功
SHI Jianming;WANG Wei;WANG Gong(Technology and Engineering Center for Space Utilization,Chinese Academy of Sciences,Beijing 100094,China)
出处
《载人航天》
CSCD
北大核心
2023年第1期102-109,共8页
Manned Spaceflight
