Research on aviation data intelligence technology based on event graph

Abstract

The new generation information technology, such as big data, artificial intelligence, has an important role in promoting the digital transformation of the aviation manufacturing industry. In view of the characteristics of multi-source and heterogeneity, few samples, and strong correlation of data in the aviation manufacturing industry, utilizing a new generation of knowledge engineering technologies such as knowledge graph with the ability of structured description and efficient management of data, a technical system and process are established in this paper for aviation data intelligence based on the event graph. The research focuses on the technical methods such as ontology modeling, event relationship recognition, event extraction, and event disambiguation for aviation data. With a selection of aviation product quality data for data intelligence technology validation, a series of work are carried out pertinent to application and prototype system developments, including classification of quality problem causes, construction of quality event graph, logic knowledge push, etc., to assist the identification of quality problem and the rapid response to quality issues. The results indicate that the technical system and path of using data and knowledge to carry out digital intelligent quality management are feasible, and the quality knowledge extraction algorithm based on event graph has strong practicality. It provides support for promoting the application of digital intelligence in the whole life cycle of aviation manufacturing industry.

Keywords

• aviation manufacturing industry
• event graph
• data intelligence
• rapid quality response