Toward Intelligent Callout and Semantic Interpretation of ISO Geometrical Production Specification

Xiangcheng Dai, Benjun Guo, Xu Yuanping, Tukun Li, Jane Jiang, Qiuyan Gai, Jia He, Jian Huang

Research output: Contribution to journalConference articlepeer-review

1 Citation (Scopus)

Abstract

ISO Geometrical Product Specification and Verification standards (GPS) is a worldwide used technique language to control the geometrical variation of a workpiece and its digital twins. More than 150 standards documents have been developed. Normally, the call out of a symbol needs to check many standards documents. It is difficult to check the complete callouts manually and lead to an increase of the specification uncertainty. To this end, this paper presents an ontology-based semantic model to graphically and integrally represent the multi-dimensional knowledge in ISO GPS. In the knowledge modelling stage, this study focuses on integrating tolerance specification and verification knowledge seamlessly. To support automatic advanced rule extractions and reasonings, this study applies Semantic Web Rule Language (SWRL) to extract and represent sufficient multi-dimensional relationships inheriting from GPS documents. Then the constraint relations and reasoners are used to support the intelligent creations and semantic interpretations of the complete tolerance specification callouts. A case study on a typical workpiece-RV reducer is undertaken to test and evaluate the validity, usability and universality of the devised knowledge model.
Original languageEnglish
Pages (from-to)42-47
Number of pages6
JournalProcedia CIRP
Volume114
DOIs
Publication statusPublished - 31 Oct 2022
Event17th CIRP Conference on Computer Aided Tolerancing - Arts et Métiers, Metz, France
Duration: 15 Jun 202217 Jun 2022
Conference number: 17
https://artsetmetiers.fr/en/CIRPCAT2022

Fingerprint

Dive into the research topics of 'Toward Intelligent Callout and Semantic Interpretation of ISO Geometrical Production Specification'. Together they form a unique fingerprint.

Cite this