Perceptual Forecasting Model of Power Big Data Based on Improved Random Forest Algorithm

Abstract

Under the background of the rapid development of the smart grid and ubiquitous power Internet of things, the number of network terminals and users has increased greatly, resulting in a gradual increase in the number and types of services that need to be carried in the distribution and consumption communication network. Although network virtualization technology can shield the differences brought by physical layer network heterogeneity, during the allocation process, the physical layer network is affected by factors such as the deployment environment, usage time, the network load, etc., making its running state time-varying. Therefore, the mapping results and transmission quality of various services are affected, and the reliability of service transmission is reduced. Based on the fact that the network operation state of the infrastructure layer directly affects the transmission quality of virtual network services, this paper introduces a reliability evaluation model based on random forest and conducts experiments and analysis on the main link mapping algorithm designed based on the evaluation model through simulation experiments. The results show that the algorithm has good resource allocation ability and a low impact on the number of services. It can further improve the acceptance rate of virtual network service mapping and improve the transmission quality of services, which is of great significance to the development of the smart grid.