Construction and application of knowledge graph for fault diagnosis of turbine generator set based on ontology. Issue 1 (1st March 2022)

• Record Type: Journal Article
• Authors: Wang, J; Yan, C F; Zhang, Y M; Li, Y J; Wang, H B
• Is Part Of: Journal of physics. Volume 2184:Issue 1(2022)
• Publication Date: 2022-03-01
• Subjects: Physics -- Congresses; 530.5
• DOI: 10.1088/1742-6596/2184/1/012015
• Languages: English
• ISSNs: 1742-6588

Abstract

Abstract: Aiming at the problems of complex structure and multi-source heterogeneity, imperfect knowledge representation, single knowledge extraction method, and difficulty of sharing and reuse in the knowledge field of turbine generator set fault diagnosis. The construction of knowledge graph is studied from multiple dimensions such as experts, fault characteristics, diagnosis techniques, research results, and solutions in the field for fault diagnosis knowledge of turbine generator set. And an ontology model of fault diagnosis knowledge for the turbine generator set is constructed. The entities, attributes, and relationships of the fault diagnosis knowledge graph for the turbine generator set are represented based on the model. The knowledge graph data are stored by the Neo4j graph database. The problems caused by multi-source and heterogeneous knowledge, fuzzy knowledge, and difficulty sharing, was solved in this field. The knowledge search system and automated quiz system based on knowledge graph are developed using the B/S framework. Many functions are realized by the knowledge graph, such as knowledge correlation, intelligent retrieval, visual display, and automated quiz, which improves the service and sharing ability of fault diagnosis knowledge for turbine generator set. Finally, the effectiveness and superiority of the system are verified by an example of a turbine generator set.