A Method for Matching Information of Substation Secondary Screen Cabinet Terminal Block Based on Artificial Intelligence

• Published in: Applied sciences Vol. 14; no. 5; p. 1904
• Main Authors: Cao, Weiguo, Chen, Zhong, Wu, Congying, Li, Tiecheng
• Format: Journal Article
• Language: English
• Published: Basel MDPI AG 01.03.2024
• Subjects: Algorithms; Artificial intelligence; branch-and-bound method; Construction; Deep learning; deep Q-network; Electricity distribution; Information storage; Labor costs; maximum common subgraph; Methods; multi-modular graph convolutional network; Network topologies; Neural networks; object detection
• ISSN: 2076-3417; 2076-3417
• DOI: 10.3390/app14051904

The matching of schematic diagrams and physical information of terminal blocks in substation secondary screen cabinets plays a crucial role in the operation and maintenance of substations. To enhance the automation level of this task and reduce labor costs, a method for identifying and matching information of terminal blocks in substation secondary screen cabinets based on artificial intelligence is investigated in this paper. Initially, multi-layer object detection networks, tailored to the characteristics of both the schematic diagrams and the physical entities in substation secondary screen cabinets, are designed for the precise extraction of information. Subsequently, network topologies for both the schematic and physical systems are established using the Neo4j database, which allows for the digital storage of information in the substation secondary screen cabinet systems. Finally, the branch-and-bound method, improved by the application of a multi-modular graph convolutional network (MGCN) and deep Q-network (DQN), is employed to solve the maximum common subgraph (MCS) problem, resulting in the rapid and efficient matching of schematic and physical data.