Scholarlink

<p>With the increasing complexity of power grid regulation scenarios, traditional monitoring systems rely on a single data source and cannot fully reflect the real-time status of the power grid, resulting in delayed fault detection and response. Therefore, this article proposes a real-time perception and intelligent warning technology for power grid regulation operations based on multidimensional data fusion. This technology integrates dynamic data from factories and stations, operation ticket content, and 3D virtual environment information, and uses deep learning and reinforcement learning algorithms to construct real-time perception and intelligent warning models. The experimental results show that the early warning time of the model in the case of short circuit fault and equipment fault is 78.2s and 66.7&#xa0;s respectively, and the processing delay in the case of equipment fault is only 70.2 ms. At the same time, the CPU utilization of the model in normal state and line overload is 47.45% and 54.87%, respectively, with low resource consumption and strong real-time and efficient performance, suitable for power grid scenarios with high requirements for fast response and efficient processing.</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

Real-time perception and intelligent early warning technology for power grid control operations based on multi-dimensional data fusion

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年7月22日

查看全文
  • Wei Li,
  • Chao Hu,
  • Bo Zhou,
  • Weibo Peng

摘要

With the increasing complexity of power grid regulation scenarios, traditional monitoring systems rely on a single data source and cannot fully reflect the real-time status of the power grid, resulting in delayed fault detection and response. Therefore, this article proposes a real-time perception and intelligent warning technology for power grid regulation operations based on multidimensional data fusion. This technology integrates dynamic data from factories and stations, operation ticket content, and 3D virtual environment information, and uses deep learning and reinforcement learning algorithms to construct real-time perception and intelligent warning models. The experimental results show that the early warning time of the model in the case of short circuit fault and equipment fault is 78.2s and 66.7 s respectively, and the processing delay in the case of equipment fault is only 70.2 ms. At the same time, the CPU utilization of the model in normal state and line overload is 47.45% and 54.87%, respectively, with low resource consumption and strong real-time and efficient performance, suitable for power grid scenarios with high requirements for fast response and efficient processing.