民航自动气象观测系统智慧应急处置平台设计与应用

Design and Application of an Intelligent Emergency Disposal Platform for Civil Aviation Automatic Weather Observation System

摘要: 针对民航自动气象观测系统设备故障处置长期依赖人工经验、流程不规范、监控与处置脱节、经验难沉淀等问题，设计并实现了一套基于B/S架构的智慧应急处置平台。平台采用Flask提供Web服务，后台线程分别执行网络设备Ping检测、MOXA串口服务器SNMP轮询和传感器状态文件解析，前端基于Canvas绘制网络拓扑。核心机制包括：基于五元组的结构化应急处置知识库；基于设备名称包含的线性匹配算法，实现告警即推送；以SQLite未解决事件列表为轻量级工作队列，实现事件创建、措施推送、执行编辑与保留/复盘；通过配置文件时间戳检测实现知识库热加载。本文还从AIOps视角简要比较了知识图谱、案例推理与机器学习方法，指出本平台适用于小规模、强实时、低资源、高可解释性场景，并讨论了跨机场移植的可行性与挑战。

Abstract: To address issues such as long-term reliance on manual experience, lack of standardized procedures, disconnection between monitoring and response, and difficulty in knowledge accumulation in handling equipment failures within civil aviation automatic meteorological observation systems, a web-based intelligent emergency response platform based on B/S architecture has been designed and implemented. The platform uses Flask to provide web services, with background threads independently performing network device Ping detection, SNMP polling of MOXA serial servers, and parsing of sensor status files, while the frontend visualizes network topology using Canvas. Key mechanisms include: a structured emergency response knowledge base based on quintuples; a linear matching algorithm based on device names enabling immediate alert push; and a lightweight task queue implemented via an SQLite unresolved event list, supporting event creation, measure distribution, execution editing, and retention/review. Knowledge base hot-loading is achieved through timestamp detection in configuration files. From an AIOps perspective, this paper briefly compares knowledge graphs, case-based reasoning, and machine learning methods, highlighting that the platform is suitable for small-scale, highly real-time, low-resource, and high-explainability scenarios, and discusses the feasibility and challenges of cross-airport deployment.