电力物联网中时延能耗均衡的MEC资源调度策略  

MEC resource scheduling strategy for delay and energy consumption balancing in Power Internet of Things

作　　者：黄东海 亢中苗 吴赞红 HUANG Donghai;KANG Zhongmiao;WU Zanhong(Power Dispatching Center,Guangdong Power Grid Company,Guangzhou Guangdong 510080,China)

机构地区：[1]广东电网有限责任公司电力调度控制中心,广东广州510080

出　　处：《太赫兹科学与电子信息学报》2024年第9期944-951,共8页Journal of Terahertz Science and Electronic Information Technology

基　　金：南方电网公司科技基金资助项目〔036000KK52220016(GDKJXM‒20220247)〕。

摘　　要：针对电力物联网中(PIoT)海量智能设备接入导致的流量激增问题,提出一种时延能耗均衡的边缘计算(MEC)资源调度策略。综合考虑信道条件、电力设备安全温度保护机制和设备能耗等因素,以兰道尔(Landaer)原理为基础构建设备侧的能耗模型和热功耗约束。在保证队列稳定性的前提下,通过联合优化任务卸载决策、传输功率和计算资源分配,最小化系统长期平均时间能耗。为解决随机优化问题,引入李雅普诺夫(Lyapunov)理论,将问题转化为每个时隙的确定性优化问题。仿真结果表明,该策略相对于基准方案能够降低系统能耗,并实现能耗与时延之间的均衡。Aiming at the traffic surge problem caused by massive smart device access in Power Internet of Things(PIoT),a resource scheduling strategy of Mobile Edge Computing(MEC)with delay and energy consumption equalization is proposed.Considering the channel conditions,the safety temperature protection mechanism of electric equipment and the energy consumption of equipment,the energy consumption model and thermal power consumption constraints on the equipment side are constructed based on the Landaer principle.Under the premise of ensuring queue stability,the long-term average time energy consumption of the system is minimized by jointly optimizing the task offloading decision,transmission power and computational resource allocation.To solve this stochastic optimization problem,Lyapunov theory is introduced to transform the problem into a deterministic optimization problem for each time slot.Simulation results show that this strategy is able to reduce the system energy consumption relative to the baseline scheme and achieve an equilibrium between energy consumption and delay.

关 键 词：电力物联网 边缘计算 任务卸载 资源分配 能耗优化 

分 类 号：TM73[电气工程—电力系统及自动化]