Intelligent Method for PDC Bit Selection Based on Graph Neural Network
Source abstract
Abstract
As oil and gas exploration extends to deep, ultra-deep, and unconventional reservoirs, high drilling costs persist. Drill bit performance, as the critical rock-breaking component, directly governs efficiency and economics. While optimal bit selection boosts rate of penetration (ROP) and cuts costs, traditional expert-dependent methods struggle to address complex formation bit parameter interactions, suffering from low accuracy and poor adaptability. With artificial intelligence gaining traction in petroleum engineering, machine learning-based bit selection has emerged as a key solution. This study focuses on polycrystalline diamond compact (PDC) bits and proposes an intelligent bit selection method based on graph neural networks (GNNs), utilizing drilling records from over 100 wells encompassing 40 multidimensional features. Through comparative analysis of four intelligent models—random forest, gradient boosting (XGBoost), gated recurrent unit (GRU), and the GNN, the results demonstrate that the GNN achieves superior performance with an R2 (coefficient of determination) of 0.932 and MAPE (mean absolute percentage error) of 6.88%. The GNN significantly outperforms conventional models in rock-breaking performance prediction. By establishing this GNN model for ROP and footage per run prediction, this study achieves intelligent bit selection that substantially enhances drilling efficiency, reduces operational costs, and provides scientifically reliable technical support for drilling operations in complex formation conditions.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Drilling and Well Engineering · Tunneling and Rock Mechanics · Hydraulic Fracturing and Reservoir Analysis
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
R&D Center for Ultra Deep Complex Reservoir Exploration and Development, CNPC, Korla 841000, China
Engineering Research Center for Ultra-Deep Complex Reservoir Exploration and Development, Xinjiang Uygur Autonomous Region, Korla 841000, China
Xinjiang Key Laboratory of Ultra-Deep Oil and Gas, Korla 841000, China
PetroChina Tarim Oilfield Company, Korla 841000, China
College of Petroleum Engineering, China University of Petroleum (Beijing), Beijing 102249, China
College of Mechanical and Transportation Engineering, China University of Petroleum (Beijing), Beijing 102249, China
References (34)
Show source-provided references
1. Research progress and the prospect of intelligent drilling and completion technologies
2. Theoretical and technological progress, challenges, and development directions of oil and gas exploration of Sinopec during the 14th Five-Year Plan period
3. Research on Bit Selection Methods for Target Formations
4. The grey clustering method is used to evaluate bit types
5. The optimization of PDC bit selection using sonic velocity profiles present in the Timor Sea
6. Spaar, J., Ledgerwood, L., Goodman, H., Graff, R., and Moo, T. (1995, January 28). Formation compressive strength estimates for predicting drillability and PDC bit selection. Proceedings of the SPE/IADC Drilling Conference and Exhibition, Amsterdam, The Netherlands.10.2523/29397-MS ↗
7. Cordy, L., Rogers, D., Karsten, D., and Davies, R. (2002, January 9). Cumulative rock strength as a quantitative means of evaluating drill bit selection and emerging PCD cutter technology. Proceedings of the IADC/SPE Asia Pacific Drilling Technology Conference and Exhibition, Jakarta, Indonesia.10.2118/77217-MS ↗
8. Fabian, R., and Birch, R. (1995, January 3–5). Canadian Application of PDC Bits Using Confined Compressive Strength Analysis. Proceedings of the CADE/CAODC Spring Drilling Conference—April, Calgary, AB, Canada.
9. Performance-Driven Drill Bit Selection Based on Field Investigations: Reducing Drilling Costs in Northern Iraq10.1016/j.rineng.2025.106163 ↗
10. Selection of Drilling Bit for the 12? Holes in the Geothermal Wells at the Company of XYZ
11. Bilgesu, H., Al-Rashidi, A., Aminian, K., and Ameri, S. (2000, January 17–19). A new approach for drill bit selection. Proceedings of the SPE Eastern Regional Meeting, Wheeling, WV, USA.10.2523/65618-MS ↗
12. Rabia, H., Farrelly, M., and Barr, M. (1986, January 3–6). A new approach to drill bit selection. Proceedings of the SPE Europec featured at EAGE Conference and Exhibition, London, UK.10.2118/15894-MS ↗
13. Bit Optimization Method and Application Based on PCA-BP
14. Jikun, Y. (2017). BP Neural Network-Based Optimization of PDC Bit Selection, China University of Petroleum (East China).
15. Bit selection method based on BP neural network and rock drillability characteristics
16. Random forests10.1023/A:1010933404324 ↗
17. Zhang, C.E. (2012). Random forests. Ensemble Machine Learning, Springer.10.1007/978-1-4419-9326-7 ↗
18. Recursive feature elimination with random forest for PTR-MS analysis of agroindustrial products10.1016/j.chemolab.2006.01.007 ↗
19. Chen, T., and Guestrin, C. (2016, January 13–17). Xgboost: A scalable tree boosting system. Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, San Francisco, CA, USA.10.1145/2939672.2939785 ↗
20. Special invited paper. additive logistic regression: A statistical view of boosting10.1214/aos/1016218223 ↗
21. Feature selection algorithm based on XGBoost
22. Cho, K., Van Merriënboer, B., Gulcehre, C., Bahdanau, D., Bougares, F., Schwenk, H., and Bengio, Y. (2014). Learning phrase representations using RNN encoder-decoder for statistical machine translation. arXiv.10.3115/v1/D14-1179 ↗
23. Enhancing short-term load forecasting with a CNN-GRU hybrid model: A comparative analysis10.1109/ACCESS.2024.3511653 ↗
24. A comprehensive survey on graph neural networks10.1109/TNNLS.2020.2978386 ↗
25. Shi, W., and Rajkumar, R. (2020, January 13–19). Point-gnn: Graph neural network for 3d object detection in a point cloud. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, Seattle, WA, USA.10.1109/CVPR42600.2020.00178 ↗
26. Advances of Adversarial Attacks and Robustness Evaluation for Graph Neural Networks
27. Frontier Advances and Applications of Graph Neural Networks
28. Myttenaere, A.D., Golden, B., Grand, B.L., and Rossi, F. (2015). Using the Mean Absolute Percentage Error for Regression Models. arXiv.10.1016/j.neucom.2015.12.114 ↗
29. An R-squared measure of goodness of fit for some common nonlinear regression models10.1016/S0304-4076(96)01818-0 ↗
30. Soper, D.S. (2021). Greed is good: Rapid hyperparameter optimization and model selection using greedy k-fold cross validation. Electronics, 10.10.3390/electronics10161973 ↗
31. Kingma, D., and Ba, J. (2014). Adam: A Method for Stochastic Optimization. arXiv.
32. Sima, Z., Tao, J., and Liu, Z. (2024, January 10–12). Adaptive and generic improvements to ResNet Backbone in image classification. Proceedings of the International Conference on Electronics, Electrical and Information Engineering (ICEEIE 2024), Bangkok, Thailand.10.1117/12.3055002 ↗
33. Llugsi, R., El Yacoubi, S., Fontaine, A., and Lupera, P. (2021, January 12–15). Comparison between Adam, AdaMax and Adam W optimizers to implement a Weather Forecast based on Neural Networks for the Andean city of Quito. Proceedings of the 2021 IEEE Fifth Ecuador Technical Chapters Meeting (ETCM), Cuenca, Ecuador.10.1109/ETCM53643.2021.9590681 ↗
34. A simple more general boxplot method for identifying outliers10.1016/j.csda.2003.10.012 ↗