Using Machine Learning Methods to Identify Coals from Drilling and Logging-While-Drilling LWD Data
- Ruizhi Zhong (The University of Queensland) | Raymond L. Johnson (The University of Queensland) | Zhongwei Chen (The University of Queensland)
- Document ID
- Unconventional Resources Technology Conference
- SPE/AAPG/SEG Asia Pacific Unconventional Resources Technology Conference, 18-19 November, Brisbane, Australia
- Publication Date
- Document Type
- Conference Paper
- 2019, Unconventional Resources Technology Conference (URTeC)
- coal seam gas (CSG), imbalanced data problem , drilling, Coal identification, machine learning
- 8 in the last 30 days
- 121 since 2007
- Show more detail
|SPE Member Price:||USD 9.50|
|SPE Non-Member Price:||USD 28.00|
Accurate coal identification is critical in coal seam gas (CSG) (also known as coalbed methane or CBM) developments because it determines well completion design and directly affects gas production. Density logging using radioactive source tools is the primary tool for coal identification, adding well trips to condition the hole and additional well costs for logging runs. In this paper, machine learning methods are applied to identify coals from drilling and logging-while-drilling (LWD) data to reduce overall well costs. Machine learning algorithms include logistic regression (LR), support vector machine (SVM), artificial neural network (ANN), random forests (RF), and extreme gradient boosting (XGBoost). The precision, recall, and F1 score are used as evaluation metrics. Because coal identification is an imbalanced data problem, the performance on the minority class (i.e., coals) is limited. To enhance the performance on coal prediction, two data manipulation techniques [naive random oversampling technique (NROS) and synthetic minority oversampling technique (SMOTE)] are separately coupled with machine learning algorithms.
Case studies are performed with data from six wells in the Surat Basin, Australia. For the first set of experiments (single well experiments), both the training data and test data are in the same well. The machine learning methods can identify coal pay zones for sections with poor or missing logs. It is found that ROP is the most important feature. The second set of experiments (multiple well experiments) use the training data from multiple nearby wells, which can predict coal pay zones in a new well. The most important feature is gamma ray. After placing slotted casings, all wells have over 90% coal identification rates and three wells have over 99% coal identification rates. This indicates that machine learning methods (either XGBoost or ANN/RF with NROS/SMOTE) can be an effective way to identify coal pay zones and reduce coring or logging costs in CSG developments.
|File Size||1 MB||Number of Pages||25|
Al-Anazi, A., and Gates, I. D. 2010. On the Capability of Support Vector Machines to Classify Lithology from Well Logs. Natural Resources Research, 19(2), 125-139. https://doi.org/10.1007/s11053-010-9118-9.
Anemangely, M., Ramezanzadeh, A., Tokhmechi, B. 2018.Drilling Rate Prediction from Petrophysical Logs and Mud Logging Data Using an Optimized Multilayer Perceptron Neural Network. Journal of Geophysics and Engineering, 15(4), 1146-1159. https://doi.org/10.1088/1742-2140/aaac5d.
Ashrafi, S. B., Anemangely, M., Sabah, M.. 2019. Application of Hybrid Artificial Neural Networks for Predicting Rate of Penetration (ROP): A Case Study from Marun Oil Field. Journal of Petroleum Science and Engineering, 175, 604-623. https://doi.org/10.1016/j.petrol.2018.12.013.
Bilgesu, H. I., Tetrick, L. T., Altmis, U.. 1997. A New Approach for the Prediction of Rate of Penetration (ROP) Values. Presented at the SPE Eastern Regional Meeting, Lexington, Kentucky, 22-24 October. SPE-39231-MS. https://doi.org/10.2118/39231-MS.
Breiman, L. 2001. Random Forests. Machine Learning, 45(1), 5-32. https://doi.org/10.1023/A:1010933404324.
Chawla, N. V., Bowyer, K. W., Hall, L. O.. 2002. SMOTE: Synthetic Minority Over-sampling Technique. Journal of Artificial Intelligence Research, 16, 321-357. https://doi.org/10.1613/jair.953.
Chawla, N. V., Japkowicz, N., and Kotcz, A. 2004. Editorial: Special Issue on Learning from Imbalanced Data Sets. SIGKDD Explor. Newsl., 6(1), 1-6. https://doi.org/10.1145/1007730.1007733.
Chen, T., and Guestrin, C. 2016. XGBoost: A Scalable Tree Boosting System. Presented at the Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, San Francisco, California, USA. https://doi.org/10.1145/2939672.2939785.
Cortes, C., and Vapnik, V. 1995. Support-Vector Networks. Machine Learning, 20(3), 273-297. https://doi.org/10.1023/A:1022627411411.
Eshkalak, M. O., Mohaghegh, S. D., and Esmaili, S. 2013. Synthetic, Geomechanical Logs for Marcellus Shale. Presented at the SPE Digital Energy Conference, The Woodlands, Texas, USA. 5-7 March. SPE-163690-MS. https://doi.org/10.2118/163690-MS.
Gidh, Y. K., Purwanto, A., and Ibrahim, H. 2012. Artificial Neural Network Drilling Parameter Optimization System Improves ROP by Predicting/Managing Bit Wear. Presented at the SPE Intelligent Energy International, Utrecht, The Netherlands. 27-29 March. SPE-149801-MS. https://doi.org/10.2118/149801-MS.
Han, H., Wang, W., and Mao, B. 2005. Borderline-SMOTE: A New Over-Sampling Method in Imbalanced Data Sets Learning. Paper presented at the Advances in Intelligent Computing, Berlin, Heidelberg. https://doi.org/10.1007/11538059_91.
Hegde, C., Daigle, H., Millwater, H.. 2017. Analysis of Rate of Penetration (ROP) Prediction in Drilling Using Physics-Based and Data-Driven Models. Journal of Petroleum Science and Engineering, 159, 295-306. https://doi.org/10.1016/j.petrol.2017.09.020.
Hegde, C., Daigle, H., and Gray, K. E. 2018. Performance Comparison of Algorithms for Real-Time Rate-of-Penetration Optimization in Drilling Using Data-Driven Models. SPE Journal, 23(05), 1706-1722. SPE-191141-PA. https://doi.org/10.2118/191141-PA.
Heinze, L., and Al-Baiyat, I. A. 2012. Implementing Artificial Neural Networks and Support Vector Machines in Stuck Pipe Prediction. Presented at the SPE Kuwait International Petroleum Conference and Exhibition, Kuwait City, Kuwait. 10-12 December. SPE-163370-MS. https://doi.org/10.2118/163370-MS.
Jolliffe, I. T., and Cadima, J. 2016. Principal Component Analysis: A Review and Recent Developments. Philosophical transactions. Series A, Mathematical, Physical, and Engineering Sciences, 374(2065), 20150202-20150202. https://doi.org/10.1098/rsta.2015.0202.
LaBelle, D., Bares, J., and Nourbakhsh, I. 2000. Material Classification by Drilling. Presented at 17th International Symposium on Automation and Robotics in Construction, Taipei, Taiwan. https://doi.org/10.22260/ISARC2000/0088.
Leung, R., and Scheding, S. 2015. Automated Coal Seam Detection Using a Modulated Specific Energy Measure in a Monitor-while-drilling Context. International Journal of Rock Mechanics and Mining Sciences, 75, 196-209. https://doi.org/10.1016/j.ijrmms.2014.10.012.
Li, G. 2010. Research on the Identification Method of Lithology Drilling with PDC Bit. Presented at the 2010 International Symposium on Computational Intelligence and Design, Hangzhou, China, 29-31 October. https://doi.org/10.1109/ISCID.2010.127.
Miller, M. S. 1980. Geophysical Logging and Exploration Techniques in the Appalachian Coal Fields. Presented at the SPE Annual Technical Conference and Exhibition, Dallas, Texas, 21-24 September. https://doi.org/10.2118/9466-MS.
Moazzeni, A., and Haffar, M. A. 2015. Artificial Intelligence for Lithology Identification through Real-Time Drilling Data. Journal of Earth Science & Climatic Change, 6(3), 1-4. https://doi.org/10.4172/2157-7617.1000265.
Ofield, B., Down, M., AI-Hashimi, L.. 2014. Efficient Open-Hole Logging of the Horizontal Wells Using Openhole Tractoring. Field Experiences From TATWEER Petroleum Company-Bahrain. Presented at the SPE Saudi Arabia Section Annual Technical Symposium and Exhibition, AI-Khobar, Saudi Arabia, 21-24 April. https://doi.org/10.2118/172172-MS.
Sabah, M., Talebkeikhah, M., Wood, D. A.. 2019. A Machine Learning Approach to Predict Drilling Rate Using Petrophysical and Mud Logging Data. Earth Science Inofrmatics, 1-21. https://doi.org/10.1007/s12145-019-00381-4.
Schmidhuber, J. 2015. Deep Learning in Neural Networks: An Overview. Neural Networks, 61, 85-117. https://doi.org/10.1016/j.neunet.2014.09.003.
Siruvuri, C., Nagarakanti, S., and Samuel, R. 2006. Stuck Pipe Prediction and Avoidance: A Convolutional Neural Network Approach. Presented at the IADC/SPE Drilling Conference, Miami, Florida, USA, 21-23 February. https://doi.org/10.2118/98378-MS.
Soares, C., and Gray, K. 2019. Real-Time Predictive Capabilities of Analytical and Machine Learning Rate of Penetration (ROP) Models. Journal of Petroleum Science and Engineering, 172, 934-959. https://doi.org/10.1016/j.petrol.2018.08.083.
Spreux, A. M., Louis, A., and Rocca, M. 1988. Logging Horizontal Wells: Field Practice for Various Techniques. Journal of Petroleum Technology, 40(10), 1352-1354. SPE-16565-PA. https://doi.org/10.2118/16565-PA.
Wang, G., and Carr, T. R. 2012. Methodology of Organic-Rich Shale Lithofacies Identification and Prediction: A Case Study from Marcellus Shale in the Appalachian Basin. Computers & Geosciences, 49, 151-163. https://doi.org/10.1016/j.cageo.2012.07.011.
Wang, K., and Zhang, L. 2008. Predicting Formation Lithology from Log Data by Using a Neural Network. Petroleum Science, 5(3), 242-246. https://doi.org/10.1007/s12182-008-0038-9.
Woods, K. S., Solka, J. L., Priebe, C. E.. 1993. Comparative Evaluation of Pattern Recognition Techniques for Detection of Microcalcifications. International Journal of Pattern Recognition and Artificial Intelligence, 7(6), 1417-1436. https://doi.org/10.1142/S0218001493000698.
Xie, Y., Zhu, C., Zhou, W.. 2018. Evaluation of Machine Learning Methods for Formation Lithology Identification: A Comparison of Tuning Processes and Model Performances. Journal of Petroleum Science and Engineering, 160, 182-193. https://doi.org/10.1016/j.petrol.2017.10.028.
Zheng, Z., Wu, X., and Srihari, R. 2004. Feature Selection for Text Categorization on Imbalanced Data. SIGKDD Explor. Newsl., 6(1), 80-89. https://doi.org/10.1145/1007730.1007741