Prediction of Carrying Capacity of Digital Twin Power Information Communication Network Based on CNN-GRU Neural Network

Abstract

With the acceleration of major national strategic deployments such as "Digital China" and "Network Power", the digital transformation of state-owned enterprises has become the general trend. In accordance with the development goals of the "14th Five-Year Plan" power communication network, the power grid company has put forward specific requirements for the digital transformation of power communication. At present, there are problems such as untimely fault response, large fault warning error and low accuracy in the power information communication network. This paper proposes a research on the carrying capacity prediction of the digital twin power information communication network based on the CNN-GRU neural network. The digital twin model collects historical and real-time carrying capacity data, predicts the carrying capacity data in different time periods through the CNN-GRU neural network, finds nodes with abnormal carrying capacity, and handles faults in time. Accurately predict the carrying capacity, so as to realize early warning of power information and communication network failures