数字孪生飞云江流域水利知识平台建设研究

Research on construction of water conservancy knowledge platform in digital twin Feiyun River Basin

摘要:

为提高飞云江流域智慧管控风险能力，根据数字孪生流域知识平台建设要求，以知识图谱为技术框架，引入机器深度学习技术，融入流域预报调度业务规则、历史典型场景洪水模式和专家预报经验等知识，通过图模型可视化处理，将知识以图数据模式进行存储，建立流域业务规则库、历史场景库、专家经验库和知识图谱库。利用图计算引擎管理和驱动知识，对流域特点、水利对象关联关系、业务规则等水利知识，进行挖掘提炼、归类组合、智慧管理和集成应用，构建基于水利知识图谱的数字孪生流域知识平台。数字孪生飞云江流域水利知识平台具有智能匹配、智慧推荐、全景推演等特点，可支撑流域防洪“四预”业务，提升流域风险与调度决策全流程数字化、智慧化、 精准化水平，为其他水利知识平台建设提供经验借鉴。

Abstract:

In order to improve the intelligent risk management and control level of Feiyun River Basin，according to the requirement of knowledge platform construction in digital twin basin，applying knowledge graph as the technical framework，machine deep learning technology is introduced，and river basin forecasting and scheduling business rules， historical typical scene fl ood patterns，and expert forecasting experience are integrated. Through visualization of the graph model，knowledge is stored as graphic data. Therefore，basin business rule database，historical scene database，expert experience database，and knowledge graph database are established. The graph computing engine is applied for knowledge management and driving，and the water conservancy knowledge such as river basin characteristics，water conservancy object association relationships，and business rules is excavated and refi ned，classifi ed and combined，intelligently managed and integrated applied. Hence，the digital twin basin knowledge platform based on water conservancy knowledge graph is constructed. The digital twin Feiyun River Basin water conservancy knowledge platform can support river basin flood control “four pre” business，with the characteristics of intelligence in matching，recommendation，and panoramic deduction. It can improve the digitalization，intelligence，and precision of the whole process of river basin risk and scheduling decision-making，serving as a reference for the construction of other water conservancy knowledge platforms.