气象业务综合监视数据标准化设计与应用

Standardization Design and Application of Integrated Monitoring Data for Meteorological Service

摘要:

打造开放的气象综合业务实时监控体系，全面提升气象业务的运行质量与效率是气象信息化发展的重要工作。针对气象综合业务监视范围广泛、内容多样以及传输密集等特性，为实现海量监视数据高效、完整及准确的采集、传输与处理，本文设计了一种基于标准化数据分类的可扩展的监视数据内容模型。该模型对种类繁多的监视数据进行抽象与概括，划分了监视数据的类别；描述了每一类别监视数据的通用属性；同时依据核心监视功能的数据需求，在通用属性的基础上进一步设计了典型的监视数据属性域内容。该模型具备通用性、可扩展性特点，在中国气象局“气象综合业务实时监控系统（天镜）”的建设中得到广泛应用，并在2022年发布为气象行业标准，为国省共有核心监控功能的实现以及各省特有监控范围的扩充提供了有效的监视数据支撑。

Abstract:

Building an open real-time monitoring system for integrated meteorological services and comprehensively improving the operational quality and efficiency of meteorological services is an important task for the development of meteorological informatization. The scope of meteorological integrated business monitoring includes multiple business fields such as forecast, prediction, observation, information, and public services. The monitoring function covers multiple aspects such as meteorological data, meteorological business applications, and infrastructure resources. The monitoring data have a wide range, diverse content, and intensive transmission, with a massive expansion in quantity and capacity. Establishing a standardised access mechanism for monitoring data has become a key factor in ensuring that complex and diverse monitoring data can be efficiently, completely, and accurately collected into the real-time monitoring system, and achieving integrated monitoring for meteorological comprehensive service. Upon the CIMISS-MCP monitoring data framework, this article designs a scalable monitoring data content model based on standardised data classification. Firstly, based on the content and application characteristics of the monitoring data, the meteorological business integrated monitoring data is divided into three categories: event information, status information, and indicator data, and their meanings are defined separately. Secondly, for each category of monitoring data, by analysing and summarising the business connotation of massive monitoring information, the necessary attribute items are extracted, forming a universal model composed of data identification domain and data attribute domain. Thirdly, considering the significant specificity of data attribute domain content due to the differences in monitoring objects and monitoring requirements, attribute items are defined for typical data domain content of event information required for core monitoring functions such as meteorological data full process monitoring, meteorological business application task monitoring, and centralised alarm monitoring. This model adopts various scalability elements in the design of general and typical attributes, which has good universality and scalability characteristics. It is widely applied in the construction of the “Meteorological Integrated Real-time Monitoring System” of the China Meteorological Administration, achieving the generation, collection, storage, calculation, service, and publication display of massive monitoring data, and well supporting various functions of the meteorological integrated monitoring business. At the same time, it is widely applied in the localisation and functional expansion of business monitoring systems in various provincial meteorological data centres and shows positive benefits. The design and application of this model effectively ensure the efficient and stable operation of meteorological information business, laying a solid data foundation for service-oriented and intelligent operation and maintenance of meteorological big data centres in the era of intelligent meteorology.