A New Energy-efficient Distribution Transformer Evaluation System Based on CensNet Graph Neural Network

As one of the core equipments of the new power distribution system, the quality of its products has a direct impact on the reliable access to distributed power sources and the supply and distribution of electricity to users. In order to improve the quality of distribution transformer products, a new energy-efficient distribution transformer evaluation system based on CensNet graph neural network is proposed. For the new energy-efficient distribution transformer, the component material and process evaluation system is firstly established from the physical level to obtain quantitative indexes; then the Neo4j graph database is used to establish a directed graph of the supply chain, to analyse the supplier roles and to obtain the graph features; finally, the CensNet graph neural network is used to evaluate the suppliers by considering the node features and edge features together.