Research On Autonomous Collaborative Decision Making For Electric Power Communication Networks Based On Digital Twins

The current power grid is developing in the direction of digital transformation,and the overall effectiveness of the communication network depends on the integrated network architecture and operation and maintenance management of the whole network.With the process of smart grid development from automation to intelligence,there are still some non-adaptations,such as inefficient distributed control,insufficient integrated service scheduling capability,and insufficient level of communication planning technology,leading to low online operation pre-decision and immediate decision making capability and intelligence level of power communication network.Therefore,it is necessary to study the key technologies of autonomous collaborative decision making of communication networks for digital transformation of power grid enterprises.This paper introduces Digital Twin technology into power communication networks.Digital Twin is an emerging technology that digitally builds physical virtual models to simulate and portray the attributes,behaviors,and rules of physical entities in the real environment.The introduction of digital twin technology in electric power communication networks can realize effective human-machine collaborative decision making,intelligent optimization,and global decision making,achieve intelligent planning of communication networks,promote the deep integration of technological innovation and traditional business in the communication field,and reduce the front-line workload at the grassroots level.In this paper,we firstly investigate the power communication network with digital twin,build the twin architecture of power communication network,introduce the process of twin network information synchronization,and construct the twin model of power communication network,which supports the subsequent research of the paper.Then,to address the problem of insufficient service integrated scheduling capability of 5G power communication network,the service autonomous scheduling method is designed,which adopts Sarsa(State Action Reward State Action)algorithm to learn the optimal access network channel selection and core network routing strategy through the SDN(Software Defined Network)controller of the twin management unit continuously interacting with the dynamic network environment.Minimize the total packet transmission delay of power communication terminals.Simulation results show that the proposed algorithm can optimize the transmission delay performance compared with Shortest Route Selection(SRS)and Maximum Channel Gain of Data Links(MaDs)algorithms.In addition,the proposed algorithms can achieve smaller fluctuations in the case of poor communication resources of some relay routers and sudden changes in wireless channel noise,which reflects the robustness in the dynamic network environment.Finally,to address the problem of insufficient level of power communication network planning technology,we study the automation technology of power communication network planning based on digital twin,propose the principle of full-cycle planning of power communication network,design the planning system architecture,and propose the method of power communication network line planning using immune algorithm solution,which can comprehensively consider the network construction cost,network reliability and other factors,and find out the optimal line planning The optimal line planning scheme is derived.The simulation results verify the effectiveness of the algorithm,and compared with the traditional genetic algorithm,the immune algorithm can avoid local optimum,and the resulting planning scheme has higher affinity and is more in line with the requirements of the power communication network.