Research On Edge Algorithm Of Multi-Area Electric Cyber-Physical System

With the development of electric cyber-phycial system(ECPS),power system security operation and control put forward a higher demand for real-time.The traditional serial algorithm and single-machine computing model can no longer meet the requirements of power system efficient processing computing tasks.Parallel distributed computing makes it possible for power systems to have high computational efficiency.A large number of intelligent power devices are used in power systems to provide a more extensive computing platform for power system computing.However,how to map different tasks to edge stations in the process of parallel and distributed computing to improve the efficiency of power system computing is a key issue to be considered.Edge computing can provide efficient and fast computing services for regional power systems.This paper analyzes three most widely used electrical operations,summarizes their common characteristics,establishes a general model of parallel and distributed computing for power systems,and forms a computing flow chart.This paper builds a task scheduling time model for edge computing,and proposes a two-stage heuristic algorithm for virtual machine configuration which is more suitable for parallel distributed computing in power system.This algorithm is compared with the traditional configuration algorithm in terms of calculation time and energy consumption.Then,the traffic interaction in this algorithm is analyzed.Compared with the traditional descending order best fit algorithm and hierarchical clustering algorithm,the results show that the proposed algorithm has better effect.At the same time,based on the network loss and deviation voltage in power system,the calculation time of traditional power flow calculation combined with three configuration algorithms is analyzed.The results show that under the same conditions,the proposed algorithm is faster and more suitable for power system parallel distributed computing.