Scholarlink

<p>With the rapid advancement of intelligent substation technologies, Digital Twin (DT) systems have become a key enabler for improving the efficiency, reliability, and intelligence of substation operation and maintenance. However, challenges persist in effectively integrating heterogeneous data sources, meeting stringent real-time performance requirements, and optimizing complex resource allocation across distributed infrastructures. To address these issues, this paper proposes an innovative cloud-edge collaborative framework for substation DT construction, grounded in differential game theory. The framework enables dynamic model evolution and real-time synchronization through distributed sensing and collaborative computation between edge nodes and cloud data centers. The cloud-based substation DT is decomposed into modular functional components, allowing for parallel processing and efficient data exchange across the cloud-edge continuum. To capture the time-varying operational dynamics of substation equipment and the heterogeneity among partial DTs, a dynamic programming formulation based on differential game theory is developed. The proposed approach jointly optimizes three critical dimensions: the strategic allocation of partial DTs from cloud data centers to edge servers, the optimal assignment of edge servers to sensor clusters for partial DT generation, and the efficient distribution of computational and communication resources for global DT integration. The optimization objective is to maximize the fidelity of the substation DT model while minimizing overall operational costs, including energy consumption and configuration overhead, over an extended operational horizon. The simulation results verify that the proposed framework significantly outperforms existing methods, achieving substantial reductions in system resource consumption and model update latency. Furthermore, it effectively satisfies the real-time performance constraints of substation operations, providing a robust theoretical and practical foundation for intelligent, sustainable substation operation and maintenance.</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 game theory-based interactive framework for digital twin substation systems

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年1月21日

查看全文
  • Yuhui Feng,
  • Chao Gao

摘要

With the rapid advancement of intelligent substation technologies, Digital Twin (DT) systems have become a key enabler for improving the efficiency, reliability, and intelligence of substation operation and maintenance. However, challenges persist in effectively integrating heterogeneous data sources, meeting stringent real-time performance requirements, and optimizing complex resource allocation across distributed infrastructures. To address these issues, this paper proposes an innovative cloud-edge collaborative framework for substation DT construction, grounded in differential game theory. The framework enables dynamic model evolution and real-time synchronization through distributed sensing and collaborative computation between edge nodes and cloud data centers. The cloud-based substation DT is decomposed into modular functional components, allowing for parallel processing and efficient data exchange across the cloud-edge continuum. To capture the time-varying operational dynamics of substation equipment and the heterogeneity among partial DTs, a dynamic programming formulation based on differential game theory is developed. The proposed approach jointly optimizes three critical dimensions: the strategic allocation of partial DTs from cloud data centers to edge servers, the optimal assignment of edge servers to sensor clusters for partial DT generation, and the efficient distribution of computational and communication resources for global DT integration. The optimization objective is to maximize the fidelity of the substation DT model while minimizing overall operational costs, including energy consumption and configuration overhead, over an extended operational horizon. The simulation results verify that the proposed framework significantly outperforms existing methods, achieving substantial reductions in system resource consumption and model update latency. Furthermore, it effectively satisfies the real-time performance constraints of substation operations, providing a robust theoretical and practical foundation for intelligent, sustainable substation operation and maintenance.