摘要
随着电力物联网和能源互联网的融合,越来越多的终端设备和用户接入配电网侧,调度云边资源为能量管理业务提供算力支持具有重要研究意义。提出了一种考虑卸载成本的配电网能量管理业务云边协同调度方法。首先介绍了能量管理业务的微服务模型和基于容器技术的云边协同业务处理架构,阐述了微服务卸载机制;接着建立了边缘节点和容器的数学模型、容器微服务队列模型和卸载成本模型,并基于边云卸载、边边卸载2种不同卸载路径分别考虑计算延时、排队延时和通信延时带来的成本,以系统整体卸载成本最小化为目标函数求解确定最优卸载路径;最后通过仿真验证了所提方法的有效性。
With the deep integration of Power Internet of Things and Energy Internet,more and more terminal devices and users are connected to the power distribution network,how to allocate cloud-side resources to provide computing support for energy management applications has great research significance.This paper proposes an offloading cost based cloud-edge collaborative scheduling method for distribution network energy management applications.Firstly,this paper introduces the microservice model of energy management applications and the cloud-side collaborative application processing structure based on docker technology,as well as expounds the microservice offloading mechanism.Then,the mathematical model of microservice,edge node and container queue are established.Based on the two different offloading paths,which are edge-cloud offloading and edge-edge offloading,the costs of computing delay,queuing delay and communication delay are considered respectively,and the optimal offloading path are determined by solving the objective function to minimize the system offloading cost.Finally,the effectiveness of the proposed method are verified by simulation.
作者
康逸群
蔡泽祥
曾兴
孙宇嫣
胡凯强
岑伯维
KANG Yiqun;CAI Zexiang;ZENG Xing;SUN Yuyan;HU Kaiqiang;CEN Bowei(School of Electric Power,South China University of Technology,Guangzhou 510640,China)
出处
《南方电网技术》
CSCD
北大核心
2021年第9期61-68,共8页
Southern Power System Technology
基金
广东省重点领域研发计划资助(2019B111109002)。
关键词
微服务卸载
卸载成本
配电网能量管理业务
云边资源优化
任务时延
microservice offloading
offloading cost
power distribution network energy management applications
cloud-side resource optimization
task delay
作者简介
康逸群(1997),女,硕士研究生,研究方向为电力物联网、边缘计算,kyiqun@163.com;蔡泽祥(1960),男,博士生导师,教授,博士,研究方向为电力系统继电保护、高压直流输电、电力物联网、边缘计算,epzxcai@scut.edu.cn;曾兴(1995),男,硕士研究生,研究方向为电力物联网、边缘计算。