Edge Computing Offload And Energy Consumption Management Strategy For Power Communication In Remote Areas

The development of distributed energy systems in remote areas has become the main direction for the current energy internet development.As an important means of data communication in the energy internet,the power communication network is slow to build in remote areas and cannot meet the computing needs of a massive distributed power source intelligent perception.In the current scenario of distributed energy business data measurement and communication in remote areas,two prominent needs are massive device access and communication demands,and low latency data perception and rapid control demands.The combination of edge computing and 5G communication technology provides real-time high-frequency perception and data computing support for the communication scenario of distributed energy business in remote areas.Existing work focuses on the communication performance optimization of the power communication access network,and the problem of the energy consumption of the power edge computing terminal in remote areas that cannot maintain long-term system operation lacks attention.Therefore,it is urgent to study the edge computing offloading and energy management strategies in the power communication scenario in remote areas.This thesis conducts research on task offloading and resource allocation strategies in the power wireless communication scenario in remote areas,with the main research contents and innovative points as follows:Firstly,in response to the communication needs of power communication and distributed power measurement business in remote areas,this thesis designs a three-layer communication architecture for remote distributed measurement,including the cloud platform scheduling layer,base station scheduling layer,and terminal equipment layer.Information transmission,resource scheduling,and collaborative computing are realized through 5G wireless public network communication technology and direct communication between devices,laying a model foundation for subsequent service caching,collaborative offloading,and energy management strategies while meeting the requirements of massive distributed intelligent terminal access and millisecond-level business communication latency.Secondly,in response to the remote and dispersed deployment of terminal equipment in remote areas and the problem that some measurement or sensing terminals cannot communicate directly with base stations and other devices,this thesis proposes a mixed communication network scheme assisted by drones for terminal equipment.A multi-hop and task offloading model between drones and measurement devices is constructed,and the task offloading strategy and multi-hop path planning problem for optimizing the total energy consumption of the system devices are established.A double-layer iterative algorithm for offloading decision-making and path planning is designed to solve the problem.Finally,in response to the limitation of communication and caching resources in the distributed power measurement communication in remote areas,this thesis considers a dynamic caching strategy that allows devices to update their service caching list based on the arrival of tasks to improve cache hit rate.A joint optimization problem for edge collaborative caching and computing strategies is established,aiming to minimize the total expected long-term energy consumption of the system while considering multiple time scales.To address the problem of multi-slot coupling and future information unavailability in the optimization problem,this thesis further proposes an online offloading and caching sharing strategy based on Lyapunov optimization.By sharing service caching content between terminals,the task computing efficiency is improved under limited energy and cache capacity.