Application research of full chain topology inference technology in the field of Nuclear Power Drawings

Under the double carbon target, with the gradual increase of China's nuclear power installed capacity and the gradual improvement of the degree of intelligence and digitalization in the industrial field, it is urgent to improve the operation efficiency of nuclear power. At present, the drawing application in the nuclear power field completely relies on manual reading of paper drawings to identify the content in the drawing, and lacks the means of intelligent application. In view of this situation, this paper proposes a set of drawing application solution based on graph theory and sorting algorithm. Based on graph theory, the topological connection relationship established by each node in the drawing is generated into a directed graph, and the generated directed graph is mapped to the external data. While the mapping relationship from the equipment entity to the directed graph is established, the connection between the equipment entity and the data is established. The calculation of the directed graph is performed using core algorithms such as depth first search, and the full link signal graph based on signal flow is derived. For the first time at home and abroad, the automatic generation of the hierarchical structure of the full link asset within or across the drawing is realized, and the digital twin of the drawing asset is constructed. At the same time, the method has strong generalization ability, which provides new ideas for the development and application of drawings in all walks of life. It creates the first intelligent application of drawings in the industrial industry, especially the nuclear power industry, and provides a foundation for the maintenance work based on machine learning, making the intelligent operation and maintenance based on drawings possible.