Research On Collaborative Scheduling And Task Offloading For Energy Internet

Recently,the Internet of Energy(Io E)which combines mobile internet technology and multi-energy technology together has become a new research hotspot gradually.Io E not only can share multiple-energy sources and achieve energy transactions,but also realize scheduling optimization of the various networks and communication resources.However,with the development and growth of networks,massive terminal devices accessing the network explosively leads to various problems such as insufficient computing resources,long network communication latency and poor system performance.Therefore,in order to deal with computing task in the Io E better,task offloading and resource collaborative scheduling are necessary to meet performance requirements including latency and energy consumption.This thesis combines mobile edge computing(MEC)technology in two scenarios of microgrid and ocean network,and studies task offloading and resource scheduling strategies from two directions of optimizing system delay and improving system utility.The main work includes:(1)For multi-user microgrid scenarios,the thesis proposes an edge-terminal collaborative task processing framework based on MEC to achieve the minimizing total delay in microgrid systems.Firstly,massive energy transaction data in the microgrid is offloaded to edge nodes or calculated locally on the device according to the offloading ratio,the computing tasks are offloaded by the partial offloading strategy.Secondly,the thesis describes the research problem as a Markov Decision Process(MDP)model,and uses the Deep Q Network(DQN)algorithm to achieve the optimal offloading decision through the interaction between the agent and the environment,so that it can realize the minimizing system latency.Finally,simulation experiments show that the performance of the DQN algorithm proposed in this thesis is superior to the two baseline algorithms,reflecting the effectiveness of the proposed algorithm.(2)For the ultra-dense edge ocean scenes composed of multiple base stations and multiple ship users,this thesis proposes an edge-cloud collaborative task processing framework based on MEC to maximize system utility.Firstly,the computing tasks of ship users are offloaded to micro-base stations or macro-base station by binary offloading strategy,and it proposes to use particle swarm optimization(PSO)algorithm to optimize transmission power allocation of ship users.Secondly,after determining the optimal transmission power of ship users,the computational resource scheduling problem is modeled as a mixed-integer nonlinear programming problem.And the thesis proposes the binary particle swarm optimization(BPSO)algorithm to maximize the system utility.Finally,simulation experiments show that the proposed PSO algorithm has better system performance and reduces energy transmission losses for power allocation optimization problems;for resource scheduling problems,the proposed BPSO algorithm outperforms existing baseline algorithms in overall performance such as system utility and response ratio,and has better stability in ultra-dense edge network scenarios at sea.