Journals →  Gornyi Zhurnal →  2025 →  #1 →  Back

MODELING OF GEOMECHANICAL PROCESSES
ArticleName AI-based geomechanical and structural core logging procedure
DOI 10.17580/gzh.2025.01.21
ArticleAuthor Selivanov D. A., Pinigin A. D., Shagitov A. M.
ArticleAuthorData

POLYUS, Moscow, Russia

D. A. Selivanov, Head of Structural Geology, selivanovda@polyus.com

 

Innopolis University, Innopolis, Russia

A. D. Pinigin, Head of AI Technologies Department
A. M. Shagitov, Data Analyst

Abstract

For structural geological and geomechanical modeling, it is critical to have complete and high-quality initial data. Amongst many types of original data, the key data is the structural and geomechanical core logs. Structural logs record cracks, faults, highly jointed zones, veins, breccia, etc. Geomechanical logging provides various indicators of rock mass quality ratings. Using these data, a geomechanical model of rock mass is built. The authors present a geomechanical and structural core logging procedure using photography and artificial intelligence. The quality of the results and their processing for the creation of a geomechanical data base was evaluated using various methods: comparison of the identified structures with the in-situ geologic logs; comparison of the obtained FF and RQD values with the in-situ geomechanical logs, with the geomechanical logs prepared by an expert, and with photographs, and also with the Priest–Hadson curve. Regarding faulting, its feature is over-identification (with some textural features identified as faults). However, this is connected with the conservative approach to filtering the results, on the basis of criticality of overlooking faults. This method embraces around 10–15 % of the data on faults. The proposed procedure allows obtaining a large bulk of data, at a high quality and within a short term. In prospect, it is planned to implement the procedure as a computer program.

keywords Core, geomechanical and structural logging, procedure, geotechnical models, artificial intelligence
References

1. Lushnikov V. N., Selivanov D. A., Berezhnoy V. P. Reliable prediction of geotechnical risks in open pit mining. Gornyi Zhurnal. 2023. No. 1. pp. 4–13.
2. Selivanov D. A. Applied structural geology for stability assessment and geotechnical risk management in mines. Gornyi Zhurnal. 2021. No. 1. pp. 54–58.
3. Trofimov A. V., Kirkin A. P., Rumyantsev A. E., Yavarov A. V. Use of numerical modelling to determine optimum overcoring parameters in rock stress–strain analysis. Tsvetnye Metally. 2020. No. 12. pp. 22–27.
4. Zhao Y., Lv W., Xu S., Wei J., Wang G. et al. DETRs Beat YOLOs on Real-time Object Detection. 2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Seattle, 2024. pp. 16965–16974.
5. Medvedev E. Yu., Voronova L. I. Identification of urban ore using RT–DETR algorithm. DSPA: Voprosy primeneniya tsifrovoy obrabotki signalov. 2024. No. 2. pp. 28–35.
6. Dosovitskiy A., Beyer L., Kolesnikov A., Weissenborn D., Zhai X. et al. An Image is worth 16×16 words: Transformers for image recognition at scale. ICLR 2020: 8th International Conference on Learning Representations. Addis Ababa, 2020.
7. Zhou Z., Siddiquee M. M. R., Tajbakhsh N., Liang J. UNet++: A Nested U-Net Architecture for medical image segmentation. Deep Learning in Medical Image Analysis and Multimodal Learning for Clinical Decision Support : Proceedings of the 4th and 8th International Workshop. Series: Lecture Notes in Computer Science. Cham : Springer, 2018. pp. 3–11.
8. Wang G., Li W., Aertsen M., Deprest J., Ourselin S. et al. Aleatoric uncertainty estimation with test-time augmentation for medical image segmentation with convolutional neural networks. Neurocomputing. 2019. Vol. 338. pp. 34–45.
9. Tereshchenko S. N., Osipov A. L., Moiseeva E. D. Detection of deer in images by computer vision methods. Siberian Journal of Life Sciences and Agriculture. 2024. Vol. 16, No. 2. pp. 431–449.
10. Barton N., Lien R., Lunde J. Engineering classification of rock masses for the design of tunnel support. Rock Mechanics. 1974. Vol. 6, Iss. 4. pp. 189–236.
11. Chen T., Kornblith S., Norouzi M., Hinton G. A simple framework for contrastive learning of visual representations. Proceedings of the 37th International Conference on Machine Learning. Vienna, 2020. Vol. 119.
12. Pikul A. S. An ensemble of modern computer vision models for deepfake detection. Bezopasnost informatsionykh tekhnologiy. 2024. Vol. 31, No. 4. pp. 116–127.
13. Ke L., Danelljan M., Li X., Tai Y.-W., Tang C.-K. et al. Mask transfiner for high-quality instance segmentation. 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition : Proceedings. New Orleans, 2022. pp. 4402–4411.

Language of full-text russian
Full content Buy
Back