Mineralogical Characterization From Geophysical Well Logs Using a Machine Learning Approach: Case Study for the Horn River Basin, Canada
Abstract Accurate estimation of mineral composition is essential for refined reservoir characterization, thermal conductivity and mechanical determinations of sedimentary rocks, but is extremely challenging in shale units due to the mineralogical complexity, low porosity and ultra‐low permeability....
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
American Geophysical Union (AGU)
2023-12-01
|
| Series: | Earth and Space Science |
| Subjects: | |
| Online Access: | https://doi.org/10.1029/2023EA003084 |
| _version_ | 1797374746100236288 |
|---|---|
| author | Kezhen Hu Xiaojun Liu Zhuoheng Chen Stephen E. Grasby |
| author_facet | Kezhen Hu Xiaojun Liu Zhuoheng Chen Stephen E. Grasby |
| author_sort | Kezhen Hu |
| collection | DOAJ |
| description | Abstract Accurate estimation of mineral composition is essential for refined reservoir characterization, thermal conductivity and mechanical determinations of sedimentary rocks, but is extremely challenging in shale units due to the mineralogical complexity, low porosity and ultra‐low permeability. Direct mineral measurements derived from laboratory X‐ray diffraction (XRD) analysis on core samples and borehole geochemical logging tool (GLT), and conventional geophysical logs from vertical wells penetrating sediments are widely available in some basins, which enables detailed mineralogical characterization of a well. A hybrid machine learning (ML) architecture that improves model training and validation by combining convolutional neural network (CNN) with XGBoost allows accurate description of the mineralogical compositions across a basin. We applied this ML approach to predict the mineral compositions using conventional well logs from the Horn River Basin, northeast British Columbia, Canada, where extensive drilling for shale‐gas and conventional hydrocarbon resources, complemented by high temperature geothermal energy potential is ideal for case testing. The predicted mineral compositions from the ML approach are consistent with the mineralogical readings from the GLT and are confirmed by the XRD mineral measurements. This allows basin‐wide mineral compositions mapping that reveals spatial trends of major mineral compositions and their relationship with the previously recognized geomechanical and geological features, which have important implications for thermal conductivity modeling, reservoir evaluation and extensive geological studies. |
| first_indexed | 2024-03-08T19:10:10Z |
| format | Article |
| id | doaj.art-f46b0e1a5e754650b7991ec39721b3e0 |
| institution | Directory Open Access Journal |
| issn | 2333-5084 |
| language | English |
| last_indexed | 2024-03-08T19:10:10Z |
| publishDate | 2023-12-01 |
| publisher | American Geophysical Union (AGU) |
| record_format | Article |
| series | Earth and Space Science |
| spelling | doaj.art-f46b0e1a5e754650b7991ec39721b3e02023-12-27T18:24:33ZengAmerican Geophysical Union (AGU)Earth and Space Science2333-50842023-12-011012n/an/a10.1029/2023EA003084Mineralogical Characterization From Geophysical Well Logs Using a Machine Learning Approach: Case Study for the Horn River Basin, CanadaKezhen Hu0Xiaojun Liu1Zhuoheng Chen2Stephen E. Grasby3Geological Survey of Canada Calgary AB CanadaGeological Survey of Canada Calgary AB CanadaGeological Survey of Canada Calgary AB CanadaGeological Survey of Canada Calgary AB CanadaAbstract Accurate estimation of mineral composition is essential for refined reservoir characterization, thermal conductivity and mechanical determinations of sedimentary rocks, but is extremely challenging in shale units due to the mineralogical complexity, low porosity and ultra‐low permeability. Direct mineral measurements derived from laboratory X‐ray diffraction (XRD) analysis on core samples and borehole geochemical logging tool (GLT), and conventional geophysical logs from vertical wells penetrating sediments are widely available in some basins, which enables detailed mineralogical characterization of a well. A hybrid machine learning (ML) architecture that improves model training and validation by combining convolutional neural network (CNN) with XGBoost allows accurate description of the mineralogical compositions across a basin. We applied this ML approach to predict the mineral compositions using conventional well logs from the Horn River Basin, northeast British Columbia, Canada, where extensive drilling for shale‐gas and conventional hydrocarbon resources, complemented by high temperature geothermal energy potential is ideal for case testing. The predicted mineral compositions from the ML approach are consistent with the mineralogical readings from the GLT and are confirmed by the XRD mineral measurements. This allows basin‐wide mineral compositions mapping that reveals spatial trends of major mineral compositions and their relationship with the previously recognized geomechanical and geological features, which have important implications for thermal conductivity modeling, reservoir evaluation and extensive geological studies.https://doi.org/10.1029/2023EA003084mineralogical characterizationgeophysical well logsmachine learningHorn River groupHorn River Basin |
| spellingShingle | Kezhen Hu Xiaojun Liu Zhuoheng Chen Stephen E. Grasby Mineralogical Characterization From Geophysical Well Logs Using a Machine Learning Approach: Case Study for the Horn River Basin, Canada Earth and Space Science mineralogical characterization geophysical well logs machine learning Horn River group Horn River Basin |
| title | Mineralogical Characterization From Geophysical Well Logs Using a Machine Learning Approach: Case Study for the Horn River Basin, Canada |
| title_full | Mineralogical Characterization From Geophysical Well Logs Using a Machine Learning Approach: Case Study for the Horn River Basin, Canada |
| title_fullStr | Mineralogical Characterization From Geophysical Well Logs Using a Machine Learning Approach: Case Study for the Horn River Basin, Canada |
| title_full_unstemmed | Mineralogical Characterization From Geophysical Well Logs Using a Machine Learning Approach: Case Study for the Horn River Basin, Canada |
| title_short | Mineralogical Characterization From Geophysical Well Logs Using a Machine Learning Approach: Case Study for the Horn River Basin, Canada |
| title_sort | mineralogical characterization from geophysical well logs using a machine learning approach case study for the horn river basin canada |
| topic | mineralogical characterization geophysical well logs machine learning Horn River group Horn River Basin |
| url | https://doi.org/10.1029/2023EA003084 |
| work_keys_str_mv | AT kezhenhu mineralogicalcharacterizationfromgeophysicalwelllogsusingamachinelearningapproachcasestudyforthehornriverbasincanada AT xiaojunliu mineralogicalcharacterizationfromgeophysicalwelllogsusingamachinelearningapproachcasestudyforthehornriverbasincanada AT zhuohengchen mineralogicalcharacterizationfromgeophysicalwelllogsusingamachinelearningapproachcasestudyforthehornriverbasincanada AT stephenegrasby mineralogicalcharacterizationfromgeophysicalwelllogsusingamachinelearningapproachcasestudyforthehornriverbasincanada |