Scholarlink

<p>With the rapid development of Artificial Intelligence (AI), big data and knowledge engineering technology, the concept of digital twins (DT) has shown great potential in refined management and intelligent dispatching of water conservancy basins. However, there are many problems in river basin system, such as multi-source heterogeneity, semantic fragmentation and non-uniform standards, which lead to the inability of efficient interconnection and intelligent collaboration of data. Therefore, this study proposes an automatic interoperability framework for heterogeneous DT basins based on semantic knowledge graph and data intermediate platform. By constructing a unified ontology model, the framework carries out semantic modeling and entity alignment for multi-source heterogeneous data, and introduces a weighted fusion algorithm to realize cross-platform data synchronization. At the semantic level, the rule-driven knowledge inference mechanism is used to complete the semantic identification and risk inference of complex events. In the execution layer, the weighted efficiency optimization model is used to balance the key indicators such as Semantic Interoperability Accuracy (SIA), Data Synchronization Latency (DSL) and Model Execution Efficiency (MEE). The experimental results show that the proposed Semantics + Intermediate Platform (G3) scheme outperforms the traditional mode (G1) and the single intermediate platform mode (G2) in multiple key indicators. Specifically, the SIA is increased by 41.2%, the DSL is reduced by 73.8%, and the MEE is shortened from 12.4&#xa0;s to 6.1&#xa0;s. Moreover, the System Scalability (SS) exhibits stronger stability under node expansion. In the cross-basin data fusion test, the fusion success rate exceeds 97% in the test scenarios, indicating that the scheme has efficient and reliable interoperability in complex basin environments. The study provides a feasible path for realizing semantic interconnection, knowledge sharing and intelligent decision-making of multi-source data in river basins, and has important reference value for smart water conservancy, ecological monitoring and DT cities.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

A semantic knowledge graph and data middleware framework for automated interoperability in digital twin watershed systems

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年5月16日

查看全文
  • Benyin Liu,
  • Xiaotao Jiang,
  • Yi Feng,
  • Zhihao Zhang

摘要

With the rapid development of Artificial Intelligence (AI), big data and knowledge engineering technology, the concept of digital twins (DT) has shown great potential in refined management and intelligent dispatching of water conservancy basins. However, there are many problems in river basin system, such as multi-source heterogeneity, semantic fragmentation and non-uniform standards, which lead to the inability of efficient interconnection and intelligent collaboration of data. Therefore, this study proposes an automatic interoperability framework for heterogeneous DT basins based on semantic knowledge graph and data intermediate platform. By constructing a unified ontology model, the framework carries out semantic modeling and entity alignment for multi-source heterogeneous data, and introduces a weighted fusion algorithm to realize cross-platform data synchronization. At the semantic level, the rule-driven knowledge inference mechanism is used to complete the semantic identification and risk inference of complex events. In the execution layer, the weighted efficiency optimization model is used to balance the key indicators such as Semantic Interoperability Accuracy (SIA), Data Synchronization Latency (DSL) and Model Execution Efficiency (MEE). The experimental results show that the proposed Semantics + Intermediate Platform (G3) scheme outperforms the traditional mode (G1) and the single intermediate platform mode (G2) in multiple key indicators. Specifically, the SIA is increased by 41.2%, the DSL is reduced by 73.8%, and the MEE is shortened from 12.4 s to 6.1 s. Moreover, the System Scalability (SS) exhibits stronger stability under node expansion. In the cross-basin data fusion test, the fusion success rate exceeds 97% in the test scenarios, indicating that the scheme has efficient and reliable interoperability in complex basin environments. The study provides a feasible path for realizing semantic interconnection, knowledge sharing and intelligent decision-making of multi-source data in river basins, and has important reference value for smart water conservancy, ecological monitoring and DT cities.