Deep learning for predicting porosity in ultra-deep fractured vuggy reservoirs from the Shunbei oilfield in Tarim Basin, China
Abstract Deep and ultra-deep carbonate reservoirs in China, which account for 34% of the country’s oil and gas reserves, pose significant challenges for porosity prediction due to their complex geological features, including extensive burial depth, weak seismic signals, and high heterogeneity. To ad...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2024-11-01
|
| Series: | Scientific Reports |
| Subjects: | |
| Online Access: | https://doi.org/10.1038/s41598-024-81051-4 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850216417921597440 |
|---|---|
| author | Ziyan Deng Dongsheng Zhou Hezheng Dong Xiaowei Huang Shiping Wei Zhijiang Kang |
| author_facet | Ziyan Deng Dongsheng Zhou Hezheng Dong Xiaowei Huang Shiping Wei Zhijiang Kang |
| author_sort | Ziyan Deng |
| collection | DOAJ |
| description | Abstract Deep and ultra-deep carbonate reservoirs in China, which account for 34% of the country’s oil and gas reserves, pose significant challenges for porosity prediction due to their complex geological features, including extensive burial depth, weak seismic signals, and high heterogeneity. To address these challenges, this study develops an advanced deep learning approach specifically designed for ultra-deep, fault-controlled, fractured-vuggy reservoirs in the Tarim Basin. The study utilizes a three-dimensional seismic dataset and applies Principal Component Analysis (PCA) to select five key features from eight seismic attributes. Additionally, seismic phase-controlled constraints are incorporated into the model. Using deep learning technology, a porosity prediction model for ultra-deep carbonate reservoirs has been constructed. Validation using blind wells from the Shunbei oilfield shows that this approach achieves a 76% reduction in Mean Square Error (MSE) compared to traditional impedance inversion techniques, highlighting its high predictive accuracy. Through SHapley Additive exPlanations (SHAP) analysis, the attributes LAMBDA_AAGFIL and PHASE_ANT are identified as the most influential, highlighting their importance in representing karst cave and fracture structures within the reservoir. These findings underscore the innovation and substantial improvement of the proposed method over conventional techniques, offering a robust and high-precision approach for porosity prediction in ultra-deep carbonate reservoirs. |
| format | Article |
| id | doaj-art-41dab2b5c3cf46b6b079ab6ff82fff89 |
| institution | OA Journals |
| issn | 2045-2322 |
| language | English |
| publishDate | 2024-11-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Scientific Reports |
| spelling | doaj-art-41dab2b5c3cf46b6b079ab6ff82fff892025-08-20T02:08:19ZengNature PortfolioScientific Reports2045-23222024-11-0114111410.1038/s41598-024-81051-4Deep learning for predicting porosity in ultra-deep fractured vuggy reservoirs from the Shunbei oilfield in Tarim Basin, ChinaZiyan Deng0Dongsheng Zhou1Hezheng Dong2Xiaowei Huang3Shiping Wei4Zhijiang Kang5Key Laboratory of Polar Geology and Marine Mineral Resources (China University of Geosciences, Beijing), Ministry of EducationKey Laboratory of Polar Geology and Marine Mineral Resources (China University of Geosciences, Beijing), Ministry of EducationKey Laboratory of Polar Geology and Marine Mineral Resources (China University of Geosciences, Beijing), Ministry of EducationKey Laboratory of Polar Geology and Marine Mineral Resources (China University of Geosciences, Beijing), Ministry of EducationKey Laboratory of Polar Geology and Marine Mineral Resources (China University of Geosciences, Beijing), Ministry of EducationPetroleum Exploration and Development Research Institute, SINOPECAbstract Deep and ultra-deep carbonate reservoirs in China, which account for 34% of the country’s oil and gas reserves, pose significant challenges for porosity prediction due to their complex geological features, including extensive burial depth, weak seismic signals, and high heterogeneity. To address these challenges, this study develops an advanced deep learning approach specifically designed for ultra-deep, fault-controlled, fractured-vuggy reservoirs in the Tarim Basin. The study utilizes a three-dimensional seismic dataset and applies Principal Component Analysis (PCA) to select five key features from eight seismic attributes. Additionally, seismic phase-controlled constraints are incorporated into the model. Using deep learning technology, a porosity prediction model for ultra-deep carbonate reservoirs has been constructed. Validation using blind wells from the Shunbei oilfield shows that this approach achieves a 76% reduction in Mean Square Error (MSE) compared to traditional impedance inversion techniques, highlighting its high predictive accuracy. Through SHapley Additive exPlanations (SHAP) analysis, the attributes LAMBDA_AAGFIL and PHASE_ANT are identified as the most influential, highlighting their importance in representing karst cave and fracture structures within the reservoir. These findings underscore the innovation and substantial improvement of the proposed method over conventional techniques, offering a robust and high-precision approach for porosity prediction in ultra-deep carbonate reservoirs.https://doi.org/10.1038/s41598-024-81051-4Tarim basinShunbeiFractured-vuggy reservoirsPorosityDeep learning |
| spellingShingle | Ziyan Deng Dongsheng Zhou Hezheng Dong Xiaowei Huang Shiping Wei Zhijiang Kang Deep learning for predicting porosity in ultra-deep fractured vuggy reservoirs from the Shunbei oilfield in Tarim Basin, China Scientific Reports Tarim basin Shunbei Fractured-vuggy reservoirs Porosity Deep learning |
| title | Deep learning for predicting porosity in ultra-deep fractured vuggy reservoirs from the Shunbei oilfield in Tarim Basin, China |
| title_full | Deep learning for predicting porosity in ultra-deep fractured vuggy reservoirs from the Shunbei oilfield in Tarim Basin, China |
| title_fullStr | Deep learning for predicting porosity in ultra-deep fractured vuggy reservoirs from the Shunbei oilfield in Tarim Basin, China |
| title_full_unstemmed | Deep learning for predicting porosity in ultra-deep fractured vuggy reservoirs from the Shunbei oilfield in Tarim Basin, China |
| title_short | Deep learning for predicting porosity in ultra-deep fractured vuggy reservoirs from the Shunbei oilfield in Tarim Basin, China |
| title_sort | deep learning for predicting porosity in ultra deep fractured vuggy reservoirs from the shunbei oilfield in tarim basin china |
| topic | Tarim basin Shunbei Fractured-vuggy reservoirs Porosity Deep learning |
| url | https://doi.org/10.1038/s41598-024-81051-4 |
| work_keys_str_mv | AT ziyandeng deeplearningforpredictingporosityinultradeepfracturedvuggyreservoirsfromtheshunbeioilfieldintarimbasinchina AT dongshengzhou deeplearningforpredictingporosityinultradeepfracturedvuggyreservoirsfromtheshunbeioilfieldintarimbasinchina AT hezhengdong deeplearningforpredictingporosityinultradeepfracturedvuggyreservoirsfromtheshunbeioilfieldintarimbasinchina AT xiaoweihuang deeplearningforpredictingporosityinultradeepfracturedvuggyreservoirsfromtheshunbeioilfieldintarimbasinchina AT shipingwei deeplearningforpredictingporosityinultradeepfracturedvuggyreservoirsfromtheshunbeioilfieldintarimbasinchina AT zhijiangkang deeplearningforpredictingporosityinultradeepfracturedvuggyreservoirsfromtheshunbeioilfieldintarimbasinchina |