Open AccessDynamic Knowledge Graph Based Construction of Quality Infrastructure System for Non-API Oil Country Tubular Goods
Author(s) -
Xiangpeng Chen,
Jing Xie,
Jianmin Gao,
Rongxi Wang,
Jian–Dong Jiang
Publication year - 2021
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/2101/1/012041
Subject(s) - computer science , knowledge graph , graph , ambiguity , quality (philosophy) , knowledge management , artificial intelligence , data mining , theoretical computer science , philosophy , epistemology , programming language
With the deep exploitation of oil and gas resources, the non-API oil country tubular goods (OCTG) adapted to specific environments are used widely. Therefore, how to effectively characterize the quality connotation of non-API OCTG to ensure their quality has become a challenge for the petroleum industry. We propose a dynamic knowledge graph of Quality Infrastructure (QI) to solve the problems of the diversity of non-API OCTG quality influencing factors, the concealment of the relationship, and the ambiguity of the mechanism of quality improvement. Firstly, a knowledge graph ontology framework of quality infrastructure is constructed, which realizes the effective combination of product characteristics and quality basic elements. Secondly, based on the professional dictionary in the field of OCTG, entity recognition adopts the entity recognition method of LDA-BiLSTM-CRF, which effectively improves the recognition accuracy of professional vocabulary. Finally, the relationship between entity types is defined as the edge of the knowledge graph; the graph embedding method is used to supplement the edge connection and calculation weight of the knowledge graph. The QI knowledge graph constructed with this technology can well describe the quality connotation of non-API OCTG, and provide opinions and methods for guaranteeing and improving the quality of non-API OCTG.