| The power Internet of Things makes full use of mobile Internet,artificial intelligence and other information technologies to realize the interconnection of everything in each link of the power system and the comprehensive state perception.The power Internet of Things based on edge computing can process computing tasks at the edge of the network and improve the quality of business service.With the edge of intelligent business development,rapid growth of electricity demand league business,single edge nodes with limited resources to support a large number of diverse demand,business request of time and space distribution further increase the edge processing load difference between nodes,need through the edge node collaboration to support the stability of the power content networking business run efficiently,so the edge nodes in the calculation of collaboration to improve power quality networking business operation object is of great significance.Edge cooperative computing takes on business processing requests through multiple edge nodes to make full use of idle node resources,improve network resource utilization,and alleviate the problem of insufficient capacity of a single node.At present,edge collaboration computing is mainly limited to the optimization of task allocation,and lacks consideration of collaboration mode and data routing optimization during collaboration,which cannot meet the differentiated needs of business.In this paper,the task assignment during the collaboration of edge nodes and the routing optimization during data transmission are studied,so as to minimize the average completion delay of the service under the condition of meeting the requirements of business resources and delay.In this paper,the Two-edge-node Cooperative-task Allocation based on Improved Particle Swarm Optimization(TCA-IPSO)is proposed to design the edge node collaboration strategy and task assignment strategy in edge collaborative computing and minimize the business completion delay while ensuring the business resources and delay requirements.Firstly,a task assignment model based on two-point collaboration is constructed.The model takes the requirements of business resources and time delay as constraints,and minimizes the average task completion time as optimization objective.Then,TCA-IPSO algorithm is proposed to solve the above models.The algorithm uses crossover and mutation operations to improve the particle updating strategy in the particle swarm optimization algorithm.While maintaining its own learning ability,the algorithm can improve the diversity of particle population and avoid premature falling into local optimum.The simulation results show that the service delay of terminal can be reduced significantly compared with the non-cooperative mode.To solve the problem of how to optimize the number of cooperative nodes in collaborative edge calculation and data routing strategy in dynamic network topology,the Improved Biogeography-based Optimization used in Task Allocation and Sending Route Optimization(IBBO-TASRO)and the Optimal Route Search based on Improved Ant Clony Optimization(ORS-IACO)were proposed.Firstly,a task assignment model based on multi-point collaboration is established,which aims to minimize the task completion delay by optimizing routing and task assignment.Then IBBO-TASRS algorithm is proposed to optimize the routing of task allocation and task dispatch,and elite retention strategy and improved migration strategy are used to avoid the biogeographic optimization algorithm falling into the local optimal solution.Finally,the ORS-IACO algorithm is proposed to optimize the routing selection in the convergence and return stage of the calculation results,and the mutation strategy in the genetic algorithm is used to improve the path selection probability in the ACO algorithm to improve the stability of the algorithm.Experimental results show that the proposed algorithm can further reduce the task completion delay and improve the quality of business operation. |
PDF Full Text Request