面向5G电力虚拟专网的确定性自适应资源管理策略
DOI:
CSTR:
作者:
作者单位:

国网福建省电力有限公司经济技术研究院

作者简介:

通讯作者:

中图分类号:

基金项目:

国网福建经研院2024年基于多类业务的5G确定性网络仿真研究(52130N240008)


Deterministic Adaptive Resource Management Strategy for 5G Electricity Virtual Private Networks
Author:
Affiliation:

福建省福州市晋安区茶园街道茶园路68号

Fund Project:

  • 摘要
  • |
  • 图/表
  • |
  • 访问统计
  • |
  • 参考文献
  • |
  • 相似文献
  • |
  • 引证文献
  • |
  • 资源附件
  • |
  • 文章评论
    摘要:

    为保障电力用户性能稳定与服务质量(QoS),提出一种面向5G电力虚拟专网的确定性自适应资源管理策略。首先,为保证电力服务资源的细粒度划分,建立5G电力虚拟专网切片系统模型,在大时间尺度上动态调整为业务保留的切片资源并建立资源约束,在小时间尺度上采用信息年龄(AoI)来保证传输时延的确定性,灵活配置功率及带宽。其次,在保证业务服务质量的前提下,提出了一个最大化长期系统效用的问题,并将该问题拆分为双时间尺度子问题。最后,为了求解资源管理最优策略,在大时间尺度上,采用双深度Q网络(DDQN)算法来做出更好的资源分配决策;在小时间尺度上,采用优先经验回放多智能体复合动作演员评论家(PER-MACA2C)算法动态调度资源。结果表明所提算法能够有效降低时延,并实现负载惩罚、成本的优化。

    Abstract:

    To ensure stable performance and Quality of Service (QoS) for power users, a deterministic adaptive resource management strategy for 5G power private networks is proposed. First, to ensure fine-grained segmentation of electric service resources, a slicing system model for 5G power virtual private network is established to dynamically adjust the slicing resources reserved for services and develop resource constraints on a large time scale, and Age of Information (AoI) is adopted on a small time scale to ensure deterministic transmission delay and flexible configuration of power and bandwidth. Second, a problem of maximizing the long-term system utility is proposed under the premise of guaranteeing the QoS, and the problem is split into dual time-scale subproblems. Finally, to solve the optimal resource management strategy, the Double Deep Q Network (DDQN) algorithm is used to make better resource allocation decisions in large time scales, and the Preferred Empirical Replay Multi-Agents Composite Action Actor Critic (PER-MACA2C) algorithm is used to dynamically schedule resources in small time scales. The results show that the proposed algorithm can effectively reduce the delay and achieve the optimization of load penalty and cost.

    参考文献
    相似文献
    引证文献
引用本文
分享
相关视频

文章指标
  • 点击次数:
  • 下载次数:
  • HTML阅读次数:
  • 引用次数:
历史
  • 收稿日期:2025-04-22
  • 最后修改日期:2025-04-22
  • 录用日期:2026-07-08
  • 在线发布日期:
  • 出版日期:
文章二维码