Prediction of Total Organic Carbon Content in Shale Based on PCA-PSO-XGBoost
Source abstract
Abstract
Total organic carbon (TOC) content is an important parameter for evaluating the abundance of organic matter in, and the hydrocarbon production capacity, of shale. Currently, no prediction method is applicable to all geological conditions, so exploring an efficient and accurate prediction method suitable for the study area is of great significance. In this study, for the shale of the Qingshankou Formation of the Gulong Sag in the Songliao Basin, TOC content prediction models using various machine learning algorithms are established and compared based on measured data, principal component analysis, and the particle swarm optimization algorithm. The results showed that GR, AC, DEN, CNL, LLS, and LLD are the most sensitive parameters using the Pearson correlation coefficient. The four principal components were also identified as input features through PCA processing. The XGBoost prediction model, established after selecting the parameters through PSO intelligence, had the highest accuracy with an R2 and RMSE of 0.90 and 0.1545, respectively, which are superior to the values of the other models. This model is suitable for the prediction of TOC content and provides effective technical support for shale oil exploration and development in the study area.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Hydrocarbon exploration and reservoir analysis · Coal Properties and Utilization · Geochemistry and Geologic Mapping
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
State Key Laboratory of Continental Shale Oil, Daqing 163318, China
Institute of Unconventional Oil & Gas Research, Northeast Petroleum University, Daqing 163318, China
School of Earth Sciences, Northeast Petroleum University, Daqing 163318, China
Exploration and Development Research Institute of Daqing Oilfield Ltd., Daqing 163712, China
Funding
National Natural Science Foundation of China
References (25)
Show source-provided references
1. Challenges and Countermeasures of Log Evaluation in Unconventional Petroleum Exploration and Development10.1016/S1876-3804(21)60089-7 ↗
2. Gao, B., Feng, Z., Luo, J., Shao, H., Bai, Y., Wang, J., Zhang, Y., Wang, Y., and Yan, M. (2024). Geochemical Characteristics of Mature to High-Maturity Shale Resources, Occurrence State of Shale Oil, and Sweet Spot Evaluation in the Qingshankou Formation, Gulong Sag, Songliao Basin. Energies, 17.10.3390/en17122877 ↗
3. Controlling Factors for Oil Production in Continental Shales: A Case Study of Cretaceous Qingshankou Formation in Songliao Basin
4. Genetic Mechanism and Petroleum Geological Significance of Calcite Veins in Organic-Rich Shales of Lacustrine Basin: A Case Study of Cretaceous Qingshankou Formation in Songliao Basin, China10.1016/S1876-3804(25)60527-1 ↗
5. Gulong Shale Oil Enrichment Mechanism and Orderly Distribution of Conventional– Unconventional Oils in the Cretaceous Qingshankou Formation, Songliao Basin, NE China10.1016/S1876-3804(23)60448-3 ↗
6. Petroleum Retention, Intraformational Migration and Segmented Accumulation within the Organic-Rich Shale in the Cretaceous Qingshankou Formation of the Gulong Sag, Songliao Basin, Northeast China10.1111/1755-6724.15066 ↗
7. Identification, Evolution and Geological Indications of Solid Bitumen in Shales: A Case Study of the First Member of Cretaceous Qingshankou Formation in Songliao Basin, NE China10.1016/S1876-3804(24)60471-4 ↗
8. Geophysical Prediction Technology for Sweet Spots of Continental Shale Oil: A Case Study of the Lianggaoshan Formation, Sichuan Basin, China10.1016/j.fuel.2024.131146 ↗
9. Artificial Intelligence Research Status and Applications in Well Logging
10. TOC Calculation of Organic Matter Rich Sediments Using Rock-Eval Pyrolysis: Critical Consideration and Insights10.1016/j.coal.2016.11.012 ↗
11. Prediction of Organic Carbon Content in Oil Shale Based on Logging: A Case Study in the Songliao Basin, Northeast China10.1007/s40948-022-00355-9 ↗
12. Researches progress in well log recognition and evaluation of source rocks
13. Prediction Model of Total Organic Carbon Content in Shale Gas
14. Comparative Study on Total Organic Carbon Content Logging Prediction Method Based on Machine Learning
15. Zhang, Y., Zhang, G., Zhao, W., Zhou, J., Li, K., and Cheng, Z. (2024). Total Organic Carbon Content Estimation for Mixed Shale Using Xgboost Method and Implication for Shale Oil Exploration. Sci. Rep., 14.10.1038/s41598-024-71504-1 ↗
16. Method and Practice of Deep Favorable Shale Reservoirs Prediction Based on Machine Learning10.1016/S1876-3804(22)60332-X ↗
17. TOC Estimation from Logging Data Using Principal Component Analysis10.1016/j.engeos.2023.100197 ↗
18. Shale Gas Production Prediction Based on PCA-PSO-LSTM Combination Model10.1142/S0218126624501767 ↗
19. On the Prediction of Geochemical Parameters (TOC, S1 and S2) by Considering Well Log Parameters Using ANFIS and LSSVM Strategies10.1016/j.petlm.2021.04.007 ↗
20. Shale Oil Production Prediction and Fracturing Optimization Based on Machine Learning10.1016/j.petrol.2022.110900 ↗
21. Prediction Method of TOC Content in Mudstone Based on Artificial Neural Network10.1088/1755-1315/781/2/022087 ↗
22. Optimization of Shale Gas Fracturing Parameters Based on Artificial Intelligence Algorithm
23. Hou, M., Xiao, Y., Lei, Z., Yang, Z., Lou, Y., and Liu, Y. (2023). Machine Learning Algorithms for Lithofacies Classification of the Gulong Shale from the Songliao Basin, China. Energies, 16.10.3390/en16062581 ↗
24. Sun, J., Dang, W., Wang, F., Nie, H., Wei, X., Li, P., Zhang, S., Feng, Y., and Li, F. (2023). Prediction of TOC Content in Organic-Rich Shale Using Machine Learning Algorithms: Comparative Study of Random Forest, Support Vector Machine, and XGBoost. Energies, 16.10.3390/en16104159 ↗
25. An Improved ΔlogR Model for Evaluating Organic Matter Abundance10.1016/j.petrol.2021.109016 ↗