Application of Machine Learning Methods in Parameter Prediction of Deep Lacustrine Oil Shale Reservoirs—Taking the Triassic Chang 7 in the Longdong Area as an Example
In the course of oil and gas exploration, understanding the petrophysical parameters such as reservoir porosity and permeability is crucial for evaluating oil and gas reserves and mining effectiveness. However, due to the complexity of well logging data and the interrelation between data, traditional analysis methods often have certain limitations and deficiencies. In order to overthrow the constraints of traditional methods, machine learning algorithms are used to build prediction models for porosity and permeability. In this study, Chang 7 reservoir in Longdong area was taken as the research object, and 320 core samples were selected for detailed measurement of porosity and permeability. In order to establish a generative prediction model, four different machine learning methods were used, including random forest (RF), K-nearest neighbor (KNN), extreme gradient boosting tree (XGBoost) and support vector machine (SVM). These methods were used to build an accurate prediction model for porosity and permeability combined with core and well logging data. In the experimental process, data preprocessing and feature selection were carried out, and four distinct machine learning methods were utilized to train and verify the model, and the optimal algorithm was selected according to the accuracy and stability of the model. Through single well analysis, the effectiveness of machine learning-based methods in predicting porosity and permeability was verified. Machine learning methods can build efficient and accurate prediction models by deeply mining information in data, providing researchers with a more detailed and comprehensive understanding of reservoir characteristics. This technological progress not only optimizes the decision-making process of reservoir development, but also improves the effectiveness of resource application, which is of great value to the advancement of the petroleum industry.
References
[1]
Abdelhamid, O., Ouafi, A., Rabah, K. et al. (2023). Integrating Drilling Parameters and Machine Learning Tools to Improve Real-Time Porosity Prediction of Multi-Zone Reservoirs. Case Study: Rhourd Chegga Oilfield, Algeria. Geoenergy Science and Engineering, 223, Article 211511. https://doi.org/10.1016/j.geoen.2023.211511
[2]
Ali, F., & Ang, R. P. (2023). Many-Dimensional Model of Adolescent School Enjoyment: A Test Using Machine Learning from Behavioral and Social-Emotional Problems. Education Sciences, 13, Article 1103. https://doi.org/10.3390/educsci13111103
[3]
Anifowose, F., Abdulraheem, A., & Al-Shuhail, A. (2019). A Parametric Study of Machine Learning Techniques in Petroleum Reservoir Permeability Prediction by Integrating Seismic Attributes and Wireline Data. Journal of Petroleum Science and Engineering, 176, 762-774. https://doi.org/10.1016/j.petrol.2019.01.110
[4]
Beijing University of Posts and Telecommunications (2020). A Method and Device for Information Reliability Evaluation Based on Knowledge Graph. CN202010245428.8.
[5]
Cui, J. F., Yang, J. L., Wang, M. et al. (2023). Shale Porosity Prediction Based on Random Forest Algorithm. Petroleum Geology and Recovery Efficiency, 30, 13-21.
[6]
Dai, L. F., Chen, S. J., Wang, P. et al. (2018). The Influence of Differences in Physical Properties of Tight Sandstone Reservoirs in the Chang-7 Section of the Ordos Basin on Oil Content. World Petroleum Industry, 1-11.
[7]
Foalem, P. L., Khomh, F., & Li, H. (2024). Studying Logging Practice in Machine Learning-Based Applications. Information and Software Technology, 170, Article 107450. https://doi.org/10.1016/j.infsof.2024.107450
[8]
Gao, F. M., Niu, K., Sun, X. P. et al. (2023). Reservoir Pore Structure Characterization and Production Prediction Based on Machine Learning. Progress in Geophysics, 1-10.
[9]
Gao, W., Jiao, C. Y., Qu, Y. L. et al. (2012). Study on the Characteristics and Diagenesis of the Heshui Tarwan Chang 7 Reservoir in the Longdong Area of the Ordos Basin. Natural Gas Exploration and Development, 35, 22-27.
[10]
Guo, G. H., Zhong, S. H., Li, S. Z. et al. (2019). Application of Machine Learning and Trace Elements of Zircon to Construct an Identification Diagram of Granite Metallogenic Potential: A Case Study of Qimantage, East Kunlun. Northwest Geology, 56, 57-70.
[11]
Guo, J. X., & Chen, M. (2018). Based on Support Vector Machine (SVM) of Turbidite Fan Sedimentary Microfacies Automatically Identify. Journal of Gansu Science, 30, 25-31.
[12]
Guo, X. L., & Gong, H. J. (2018). Application of Entropy Weight Fuzzy Comprehensive Evaluation in Reservoir Evaluation. Journal of Shaanxi University of Technology (Natural Science Edition), 34, 21-27.
[13]
Han, W. X., Tao, S. Z., Yao, J. L. et al. (2016). Fine Characterization of Tight Reservoirs in the Longdong Area of the Ordos Basin. Natural Gas Geosciences, 27, 820-826.
[14]
Hofmann, J., Li, Z., Taphorn, K., Herzen, J., & Wudy, K. (2024). Porosity Prediction in Laser-Based Powder Bed Fusion of Polyamide 12 Using Infrared Thermography and Machine Learning. Additive Manufacturing, 85, Article 104176. https://doi.org/10.1016/j.addma.2024.104176
[15]
Hou, K. J., Wu, J. M., Ge, X. et al. (2019). Calculation Method of Reservoir Porosity in Expanded Section Based on Two-Dimensional Nuclear Magnetic Data. In Proceedings of the 31st National Natural Gas Academic Conference (Geological Exploration) (p. 8). Sinopec Southwest Petroleum Engineering Ltd.
[16]
Hou, X. M., Wang, F. Y., Zai, Y. et al. (2022) Based on Machine Learning and Log Data of Carbonate Porosity and Permeability Prediction. Journal of Jilin University (Earth Sciences), 52, 644-653.
[17]
Huang, Q. L., Zhao, J. L., Bai, Q. et al. (2023). Automatic Identification of Sedimentary Microfacies Based on Adaboost Algorithm: A Case Study of Shanxi Formation, Q Area, Longdong Gas Field. Geological Bulletin, 43, 658-666.
[18]
Huang, X. J. (2019). Simulation and Optimization of Ground Source Heat Pump System Based on TRNSYS. Suzhou University of Science and Technology.
[19]
Hubei Seismological Bureau (Institute of Seismology, China Earthquake Administration) (2023). A Method and System for Pre-Earthquake Gravity Disturbance Signal Recognition. CN202310089610.2.
[20]
Kumar, P. S., David, L., & Vikram, V. (2023). A Robust Mechanistic Model for Pore Pressure Prediction from Petrophysical Logs Aided by Machine Learning in the Gas Hydrate-Bearing Sediments over the Offshore Krishna-Godavari Basin, India. Natural Resources Research, 32, 2727-2752.
[21]
Li, L. (2013). Research on Micromagnetic Detection and Imaging Techniques of Crystalline Silicon Defects. Nanchang Hangkong University.
[22]
Liang, X., Wang, S., Zhou, M. S. et al. (2017). Coal Reservoir Porosity Evaluation Based on Nuclear Magnetic Resonance and Resistivity Logging. Coal Engineering, 49, 130-133.
[23]
Liu, H., Xu, J. X., Chen, H. B. et al. (2023) Porosity Calculation Method for Shale Gas Reservoirs Based on Nuclear Magnetic Resonance Logging Correction. China Offshore Oil and Gas, 35, 89-95.
[24]
Liu, J., Tian, L., Liu, S. X. et al. (2023). Tight Gas Well Productivity Prediction Model Based on Composite Machine Algorithm: A Case Study of SM Block in Ordos Basin. Petroleum Geology and Development in Daqing, 43, 69-78.
[25]
Liu, K., Niri, M. F., Apachitei, G., Lain, M., Greenwood, D., & Marco, J. (2022). Interpretable Machine Learning for Battery Capacities Prediction and Coating Parameters Analysis. Control Engineering Practice, 124, Article 105202. https://doi.org/10.1016/j.conengprac.2022.105202
[26]
Liu, T., & Zhang, R. (2024). A Machine Learning-Based Hybrid Model for Fracture Parameterization and Distribution Prediction in Unconventional Reservoirs. Computers and Geotechnics, 168, Article 106146. https://doi.org/10.1016/j.compgeo.2024.106146
[27]
Luo, C. Q. (2023). Research on PCB Assembling Scheme Optimization and Material Prediction Method under Multiple Constraints. Chongqing University of Technology.
[28]
Luo, G., Xiao, L. Z., Shi, Y. Q. et al. (2022). Research on Fluid Identification Method of Tight Reservoir Based on Machine Learning. Bulletin of Petroleum Science, 7, 24-33.
[29]
Ma, D. B., Li, M., Cui, W. J. et al. (2010). Distribution Pattern of Turbidite Rocks in the Chang 6-Chang 7 Sections of the Yanchang Formation in Longdong Area. Xinjiang Petroleum Geology, 31, 33-36.
[30]
Martin, E., Samuel, P., Max, O. et al. (2024). Analysis of Data Generation and Preparation for Porosity Prediction in Cold Spray Using Machine Learning. Journal of Thermal Spray Technology,33, 1270-1291.
[31]
Pan, S. W., Zheng, Z. C., Lei, J. Y. et al. (2019). Sandstone Reservoir Porosity Prediction Based on Hybrid Optimization XGBoost Algorithm. Computer Applications and Software, 40, 103-109+206.
[32]
Qu, H. J., Pu, R. H., Chen, S. et al. (2019). Facies and Potential Coupling Control of Mesozoic Oil Accumulation in Ordos Basin. Oil & Gas Geology, 40, 752-762+874.
[33]
Ren, J., Zhai, F. F., Li, F. L. et al. (2017). Based on The Natural Potential Method of Argillaceous Sandstone Reservoir Porosity Calculation. Journal of Logging Technology, 9, 292-295.
[34]
Shan, X. G. (2014). Identification and Potential Evaluation of Chang 6 Thin Reservoir in Yanchang Formation of Qilicun Oilfield. Xi’an Shiyou University.
[35]
She, G., Gui, P. F., Chen, Y. et al. (2019). Ping West Qaidam Basin Region Bedrock Reservoir Logging Evaluation. Journal of Yangtze University (Natural Science Edition), 9, 18-26.
[36]
Shi, P. Y., Wang, C. Y., Yan, W. L. et al. (2019). Metamorphic Mineral Content Was Calculated by the Logging Data Element Method. Journal of Logging Technology, 6, 597-600.
[37]
Sima, L. Q., Wang, P. C., Deng, X. H. et al. (2008). Discussion on Reservoir Porosity Calculation Method of Feixianguan Formation in Northeast Sichuan. Journal of Southwest Petroleum University (Natural Science Edition), 2, 1-4+183.
[38]
Sun, B. (2016). Research on the Method of Calculating Porosity of Tight Sandstone Formations Using Acoustic Logging Curves. Petrochemical Technology, 23, 163.
[39]
Wang, D. W., Li, J. Y., Liu, D. et al. (2024). Characterization of Capillary Pressure Curve and Pore Throat Distribution Based on Reservoir Physical Parameters. Unconventional Oil and Gas, 11, 1-9.
[40]
Wang, L. J. (2023). Reservoir Porosity Prediction Based on Deep Learning. Qingdao University of Technology.
[41]
Wang, X. G., Fang, Y., Fang, J. et al. (2015). Discussion on Calculation Method of Matrix Porosity in Limestone Reservoir. Petroleum Geology and Engineering, 29, 81-83.
[42]
Wang, Z. H., Zhang, C. G., Chai, C. Y. et al. (2004). Logging Identification Model of Low Permeability Reservoir Types. Natural Gas Industry, 9, 36-38.
[43]
Wang, Z. L. (2016). Calculation and Application of Complex Lithologic Reservoir Parameters of Chang 6 Formation in Zhoujiawan area, Changqing Oilfield. China University of Petroleum.
[44]
Wen, Z. G., Luo, Y. S., Liu, J. Y. et al. (2022). Pore Structure Characteristics and Genesis Mechanism of the Triassic Chang 7 Shale Oil Reservoir in Longdong Area. Lithological Oil and Gas Reservoir, 34, 47-59.
[45]
Zhan, Y. X., Chen, L., Lin, C. H. et al. (2021). Determination of Reservoir Formation Stages and Periods of Chang-7 Shale Oil in Longdong Area of Ordos Basin. Yunnan Chemical Industry, 48, 121-125.
[46]
Zhang, P., & Zhang, X.L. (2014). Research on Permeability Logging Interpretation Model of Low Porosity and Low Permeability Reservoir. Groundwater, 36, 74-76.
[47]
Zhang, Q. (2019). Study on the Microstructural Characteristics of the Tight Sandstone Reservoir of Group 7 in the Triassic Yanchang Formation in the LongdongArea of the Ordos Basin. Northwest University.
[48]
Zhang, Y. H., Shi, B. H., Zhang, Y. J. et al. (2019). Application of Machine Learning Method to Prediction of Shallow Beach and Bar Facies Thin Reservoir Porosity: A Case Study of Cretaceous in Chepaizi Area, Junggar Basin. Acta Sedimentologica Sinica, 41, 1559-1567.
[49]
Zhang, Z. Q. (2023). Research on the Identification Method of Fluid Properties of Pyroclastic Sandstone Conglomerate Reservoir Based on Machine Learning. Northeast Petroleum University.
[50]
Zhao, J., Wang, Q., Rong, W., Zeng, J., Ren, Y., & Chen, H. (2024). Permeability Prediction of Carbonate Reservoir Based on Nuclear Magnetic Resonance (NMR) Logging and Machine Learning. Energies, 17, Article 1458. https://doi.org/10.3390/en17061458
