Intelligent Routing Distribution Algorithm For Dynamic Business Needs Of Power Internet Of Things

In the context of large-scale interconnection of the power grid,the power grid has introduced the Internet of Things technology to form a new type of power Internet of Things,which enables the power grid to have new features such as comprehensive perception,extensive connectivity,and open sharing.The businesses supported by the power Internet of Things mainly include integrated energy trading,distributed energy management and trading,demand response,energy storage management,electric vehicle networking,virtual power plant power trading,and UAV inspections of transmission lines.In order to meet the needs of intelligent and real-time operation and control of the power system,the Internet of Things needs to be able to predict the bandwidth requirements of the business at a certain time in the future in real time,and at the same time reasonably allocate the business bandwidth resources,and on this basis,the business Carry out efficient and reliable routing planning.Existing traffic forecasting and bandwidth allocation technologies are insufficient in coping with a surge or decrease in traffic,which may result in insufficient or excessive allocation of bandwidth.In addition,the existing routing planning technology also lacks effective consideration of the overall grid outage risk.In response to the above problems,this article focuses on the needs of power Internet of Things,focusing on the real-time traffic forecasting and bandwidth allocation methods of communication network services,and the comprehensive risk-driven routing planning method to ensure the reliable and safe operation of the power grid.Improve the operating efficiency of the power Internet of Things.First,the real-time traffic prediction and bandwidth allocation of the power Internet of Things are studied.By modeling the long-term traffic change characteristics of dynamic business traffic,and introducing the real-time traffic change characteristics of the business,combining these two points,it proposes a GRU with AUGRU(GRU with Attention Update Gate(AUGRU)real-time service traffic prediction model,and bandwidth compensation method based on real-time traffic.Specifically,real-time service flow information is used to correct the predicted bandwidth,which improves the accuracy of real-time service flow prediction.After that,based on the predicted bandwidth of the business above,a routing planning method for comprehensive risk-driven power Internet of Things business was proposed.This method first establishes a comprehensive risk assessment model for the power grid.In the risk assessment model,the node load balance degree,the link load balance degree,and the average delay of the service are considered.At the same time,rigid requirements such as the upper limit of the load of the node and the link,and the upper limit of the maximum delay of the service are added.Secondly,the comprehensive risk assessment value is used as the reward value of the DQN(Deep Q-Learning,DQN)model,and routing planning is performed through the DQN model under a fixed bandwidth,and the routing planning method with the smallest comprehensive risk value is solved.After further considering the dynamic bandwidth requirements of the business,a DDQN(Double Deep Q-learning,DDQN)model based on the bandwidth allocation and routing planning of the business was proposed to ensure the real-time bandwidth demand of the business.Finally,simulation experiments verify the effectiveness and practical value of the method proposed in this paper.Through experimental verification,the method proposed in this article can predict business traffic in real time and accurately,and efficiently allocate bandwidth for the business,and at the same time,based on the predicted business traffic and allocated bandwidth,assign a route with the least comprehensive risk to the business.So as to effectively guarantee the reliable and stable operation of the power system.