DESIGN AND MODELING

Formal verification of workpiece locating datum sets from a STEP model using a geometric constraint matrix

Authors

  • Marina O. Zhuravleva https://orcid.org/0009-0001-9282-2807 MIREA – Russian Technological University, 119454, Moscow, Vernadskogo Avenue, 78
  • Andrey D. Nazarov MIREA – Russian Technological University, 119454, Moscow, Vernadskogo Avenue, 78
  • Ivan V. Grigorev MIREA – Russian Technological University, 119454, Moscow, Vernadskogo Avenue, 78

How to cite

GOST Zhuravleva M. O., Nazarov A. D., Grigorev I. V. Formal verification of workpiece locating datum sets from a STEP model using a geometric constraint matrix // STROITEL'NYE I DOROZHNYE MASHINY. 2026. Vol. 70. No. 10. P. 140-151. DOI: 10.25726/p4303-1009-1079-y
APA Zhuravleva, M. O., Nazarov, A. D. & Grigorev, I. V. (2026). Formal verification of workpiece locating datum sets from a STEP model using a geometric constraint matrix. STROITEL'NYE I DOROZHNYE MASHINY, 70(10), 140-151. https://doi.org/10.25726/p4303-1009-1079-y

Abstract

The study formalizes verification of technological datum sets from the boundary representation (B-Rep) of a STEP model and separates kinematic admissibility of a set from statistical ranking of faces. Three tasks were solved: translating GOST 21495-76 provisions into verifiable software rules; building a geometric constraint matrix in which a support point is a row [n, r × n], the rank gives the number of constrained degrees of freedom and the null space gives retained motions; and evaluating the reproducibility of expert annotation. On 48 synthetic STEP models from eight families the required rank was obtained for 54 of 54 reference sets, and all 48 negative sets were rejected with the specified status. Leave-one-family-out macro F1 was 0.333 for geometric rules, 0.416 for a multilayer perceptron and 0.465 for the hybrid mode. Two manufacturing engineers independently labelled 200 faces of six external models, the second one blind; all 200 labels (Cohen's κ = 1.000) and 6 of 6 datum sets matched. A pilot classifier trained on expert labels correctly identified 7 of 11 faces of the held-out part. The method can be used for preliminary verification of locating schemes in process planning: the report gives the number of independent constraints, free motions and the clause of the standard behind each decision.

Keywords

technological datums workpiece location GOST 21495-76 STEP model geometric constraint matrix degrees of freedom expert annotation machine learning

Funding

The authors did not declare any external funding for this research.

References

Bazrov B.M., Gainutdinov T.M. Determination of the Minimum Number of Technological Bases of a Part // Russian Engineering Research. 2020. no. 9. pp. 73-75.

GOST 21495-76. Basing and Bases in Mechanical Engineering. Terms and Definitions. Introduced 1977-01-01. Moscow: Izdatel'stvo Standartov, 1990. 35 pp.

Cao W., Robinson T., Hua Y., Boussuge F., Colligan A.R., Pan W. Graph representation of 3D CAD models for machining feature recognition with deep learning // Proceedings of the ASME 2020 International Design Engineering Technical Conferences and Computers and Information in Engineering Conference. 2020. Vol. 11A. Art. V11AT11A003.

Colligan A.R., Robinson T.T., Nolan D.C., Hua Y., Cao W. Hierarchical CADNet: learning from B-Reps for machining feature recognition // Computer-Aided Design. 2022. Vol. 147. Art. 103226.

Ding D., Liu Y.H., Wang M.Y., Wang S. Automatic selection of fixturing surfaces and fixturing points for polyhedral workpieces // IEEE Transactions on Robotics and Automation. 2001. Vol. 17. no. 6. pp. 833-841.

Fu X., Peddireddy D., Zhou F., Xi Y., Aggarwal V., Li X., Jun M.B.-G. Boundary representation compatible feature recognition for manufacturing CAD models // Manufacturing Letters. 2023. Vol. 35. Suppl. pp. 895-903.

Jayaraman P.K., Sanghi A., Lambourne J.G., Willis K.D.D., Davies T., Shayani H., Morris N. UV-Net: learning from boundary representations // Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. 2021. pp. 11703-11712.

Koch S., Matveev A., Jiang Z., Williams F., Artemov A., Burnaev E., Alexa M., Zorin D., Panozzo D. ABC: a big CAD model dataset for geometric deep learning // Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. 2019. pp. 9601-9611.

Lambourne J.G., Willis K.D.D., Jayaraman P.K., Sanghi A., Meltzer P., Shayani H. BRepNet: a topological message passing system for solid models // Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. 2021. pp. 12773-12782.

Lee J., Yeo C., Cheon S.-U., Park J.H., Mun D. BRepGAT: graph neural network to segment machining feature faces in a B-rep model // Journal of Computational Design and Engineering. 2023. Vol. 10. no. 6. pp. 2384-2400.

Liu X., Yang H., Wang Y., Qiu L., Zhang S., Tan J. A workpiece setup optimization method for 5-axis machining with motion coherence and stiffness enhancement // Precision Engineering. 2024. Vol. 88. pp. 867-883.

Mei J., Hong-C.Z., Oldham W.J.B. A neural network approach for datum selection in computer-aided process planning // Computers in Industry. 1995. Vol. 27. no. 1. pp. 53-64.

Samsonov V., Chrismarie E., Köpken H.-G., Bär S., Lütticke D., Meisen T. Deep representation learning and reinforcement learning for workpiece setup optimization in CNC milling // Production Engineering. 2023. Vol. 17. no. 6. pp. 847-859.

Wakhare M., Šormaz D.N., Fuloria M. Artificial intelligence-based setup planning expert system for 3-axis and 4-axis machining for Industry 4.0 // The International Journal of Advanced Manufacturing Technology. 2026. Vol. 144. no. 9-10. pp. 7461-7486.

Wu H., Lei R., Peng Y., Gao L. AAGNet: a graph neural network towards multi-task machining feature recognition // Robotics and Computer-Integrated Manufacturing. 2024. Vol. 86. Art. 102661.

Xia M., Zhao X., Hu X. Machining feature and topological relationship recognition based on a multi-task graph neural network // Advanced Engineering Informatics. 2024. Vol. 62. Part B. Art. 102721.

Zhang S., Guan Z., Jiang H., Wang X., Tan P. BrepMFR: enhancing machining feature recognition in B-rep models through deep learning and domain adaptation // Computer Aided Geometric Design. 2024. Vol. 111. Art. 102318.

Zhang Y., Yu X., Sun J., Zhang Y., Xu X., Gong Y. Intelligent STEP-NC-compliant setup planning method // Journal of Manufacturing Systems. 2022. Vol. 62. pp. 62-75.

Metrics

13 views
0 downloads
Want to publish with us?
Submit an article

Similar Articles

1 2 3 4 5 > >> 

You may also start an advanced similarity search for this article.