Construction of Power System Communication Network Based on Edge Computing

In order to overcome the problem of network congestion in the traditional cloud computing mode of power system, the author introduces edge computing technology on the basis of the existing power system communication network. On the basis of analyzing the large data computing capacity of the domestic power system, the optimization approaches for the existing communication network architecture of the power system were discussed. We have constructed an overall model of cloud edge collaborative power system communication network based on Docker technology. Apply collaborative operation mechanism to all levels of equipment in the system. Based on the characteristics of the communication network model in the power system, a global optimization scheme for task collaboration mechanism based on cooperative coevolution was studied using multiple unmanned aerial vehicle inspection tasks as the application scenario. A delay model of edge computing server is proposed. Discussed a computing offloading strategy based on the comprehensive optimization of latency and resource allocation. Constructed a resource allocation objective function. The author uses edge computing technology to quickly determine the location of abnormal traffic attacks and realize rapid detection of abnormal traffic attacks. The experimental results show that the average error of attack location detection obtained by the two comparative detection methods is 3.2 m and 2.4 m, respectively, and the average error of attack location detection obtained by the optimized design detection method is 0.4 m. It has been proven that the minimum difference in resource allocation is achieved while ensuring the minimum total delay of each node.