Scholarlink

<p>Real-time data acquisition and processing are crucial for intelligent-command-network interactive systems, as data delays, losses, or errors may degrade system performance. This study presents the design and application of an interactive system for intelligent power-grid command networks that integrates digital twins and MATLAB. The system architecture comprises perception, network, and application layers. A nearest-neighbor algorithm is introduced to classify command data while dynamically updating cluster centers. A support vector machine (SVM) serves as the data analysis model to detect anomalies in power grid data. Using digital twin mapping methodology, an improved YOLOv4 lightweight model is designed to achieve spatial mapping and state updates for power grid equipment twin models. An interactive data load analysis model for intelligent command networks is constructed based on a wavelet neural network (WNN) algorithm, and MATLAB-based web functions are implemented for remote network data interaction. The experimental results demonstrate that the proposed method's loss value progressively decreases, the F1-score exhibits minimal fluctuation across different device types, the analytical output aligns closely with expected values, and network interactive data retrieval is completed within 3&#xa0;min.</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

Design and application of interactive system of intelligent power-grid command network based on digital twin and MATLAB

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年10月2日

查看全文
  • Liwei Wang,
  • Xue Li,
  • Xuan Guo

摘要

Real-time data acquisition and processing are crucial for intelligent-command-network interactive systems, as data delays, losses, or errors may degrade system performance. This study presents the design and application of an interactive system for intelligent power-grid command networks that integrates digital twins and MATLAB. The system architecture comprises perception, network, and application layers. A nearest-neighbor algorithm is introduced to classify command data while dynamically updating cluster centers. A support vector machine (SVM) serves as the data analysis model to detect anomalies in power grid data. Using digital twin mapping methodology, an improved YOLOv4 lightweight model is designed to achieve spatial mapping and state updates for power grid equipment twin models. An interactive data load analysis model for intelligent command networks is constructed based on a wavelet neural network (WNN) algorithm, and MATLAB-based web functions are implemented for remote network data interaction. The experimental results demonstrate that the proposed method's loss value progressively decreases, the F1-score exhibits minimal fluctuation across different device types, the analytical output aligns closely with expected values, and network interactive data retrieval is completed within 3 min.