基于边缘代理及深度学习的轻量型云边协同框架研究

Research of Lightweight Cloud Edge Collaboration Framework Based on Edge Agent and Deep Learning

  • 摘要: 随着电网建设的发展,配电站房及其设备规模持续增长,对边缘物联代理的实用化提出了更多挑战。在Kubernetes常规容器化架构的基础上,提出轻量化设计思路减少边缘物联代理装置对硬件资源的占用和消耗,实现云−边−端业务协同、数据协同和智能协同,同时在边缘侧开展了业务应用APP以及AI视觉学习模型算法库的规划设计,较好地满足了配电网运检业务的需求。

     

    Abstract: With the development of power grid construction and the continuous growth of the scale of distribution station buildings and equipment, in order to support the development of power grid business applications, more challenges have been put forward to the practicality of edge Internet of things (IOT) agents, and the conventional containerization technology route has appeared a certain bottleneck. Through the introduction and selection of containerization technology, this paper puts forward the lightweight design idea based on the conventional containerization architecture of Kubernetes. The research results of this paper greatly reduce the occupation and consumption of hardware resources by the edge IOT agent device, and realize cloud edge end business collaboration, data collaboration and intelligent collaboration. At the same time, the planning and design of business application app and AI visual learning model algorithm library are carried out on the edge side, which better meets the support of distribution network operation inspection business.

     

/

返回文章
返回