Optimization of Offshore Saline Aquifer CO<sub>2</sub> Storage in Smeaheia Using Surrogate Reservoir Models
Machine learning-based Surrogate Reservoir Models (SRMs) can replace/augment multi-physics numerical simulations by replicating the reservoir simulation results with reduced computational effort while maintaining accuracy compared with numerical simulations. This research will demonstrate SRMs’ pote...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2024-10-01
|
| Series: | Algorithms |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1999-4893/17/10/452 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850204975739699200 |
|---|---|
| author | Behzad Amiri Ashkan Jahanbani Ghahfarokhi Vera Rocca Cuthbert Shang Wui Ng |
| author_facet | Behzad Amiri Ashkan Jahanbani Ghahfarokhi Vera Rocca Cuthbert Shang Wui Ng |
| author_sort | Behzad Amiri |
| collection | DOAJ |
| description | Machine learning-based Surrogate Reservoir Models (SRMs) can replace/augment multi-physics numerical simulations by replicating the reservoir simulation results with reduced computational effort while maintaining accuracy compared with numerical simulations. This research will demonstrate SRMs’ potential in long-term simulations and optimization of geological carbon storage in a real-world geological setting and address challenges in big data curation and model training. The present study focuses on CO<sub>2</sub> storage in the Smeaheia saline aquifer. Two SRMs were created using Deep Neural Networks (DNNs) to predict CO<sub>2</sub> saturation and pressure over all grid blocks for 50 years. 18 million samples and 31 features, including reservoir static and dynamic properties, build the input data. Models comprise 3–5 hidden layers with 128–512 units apiece. SRMs showed a runtime improvement of 300 times and an accuracy of 99% compared to the 3D numerical simulator. The genetic algorithm was then employed to determine the optimal rate and duration of CO<sub>2</sub> injection, which maximizes the volume of injected CO<sub>2</sub> while ensuring storage operations’ safety through constraints. The optimization continued for the reproduction of 100 generations, each containing 100 individuals, without any hyperparameter tuning. Finally, the optimization results confirm the significant potential of Smeaheia for storing 170 Mt CO<sub>2</sub>. |
| format | Article |
| id | doaj-art-c3f07c88e45444a3883a10de4feecb78 |
| institution | OA Journals |
| issn | 1999-4893 |
| language | English |
| publishDate | 2024-10-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Algorithms |
| spelling | doaj-art-c3f07c88e45444a3883a10de4feecb782025-08-20T02:11:11ZengMDPI AGAlgorithms1999-48932024-10-01171045210.3390/a17100452Optimization of Offshore Saline Aquifer CO<sub>2</sub> Storage in Smeaheia Using Surrogate Reservoir ModelsBehzad Amiri0Ashkan Jahanbani Ghahfarokhi1Vera Rocca2Cuthbert Shang Wui Ng3Department of Energy Resources, University of Stavanger, 4021 Stavanger, NorwayDepartment of Geosciences, Norwegian University of Science and Technology, 7031 Trondheim, NorwayDepartment of Environment, Land and Infrastructure Engineering, Politecnico di Torino, 10129 Torino, ItalyDepartment of Geosciences, Norwegian University of Science and Technology, 7031 Trondheim, NorwayMachine learning-based Surrogate Reservoir Models (SRMs) can replace/augment multi-physics numerical simulations by replicating the reservoir simulation results with reduced computational effort while maintaining accuracy compared with numerical simulations. This research will demonstrate SRMs’ potential in long-term simulations and optimization of geological carbon storage in a real-world geological setting and address challenges in big data curation and model training. The present study focuses on CO<sub>2</sub> storage in the Smeaheia saline aquifer. Two SRMs were created using Deep Neural Networks (DNNs) to predict CO<sub>2</sub> saturation and pressure over all grid blocks for 50 years. 18 million samples and 31 features, including reservoir static and dynamic properties, build the input data. Models comprise 3–5 hidden layers with 128–512 units apiece. SRMs showed a runtime improvement of 300 times and an accuracy of 99% compared to the 3D numerical simulator. The genetic algorithm was then employed to determine the optimal rate and duration of CO<sub>2</sub> injection, which maximizes the volume of injected CO<sub>2</sub> while ensuring storage operations’ safety through constraints. The optimization continued for the reproduction of 100 generations, each containing 100 individuals, without any hyperparameter tuning. Finally, the optimization results confirm the significant potential of Smeaheia for storing 170 Mt CO<sub>2</sub>.https://www.mdpi.com/1999-4893/17/10/452geological carbon storagesurrogate reservoir modelartificial intelligencedeep learningoptimization |
| spellingShingle | Behzad Amiri Ashkan Jahanbani Ghahfarokhi Vera Rocca Cuthbert Shang Wui Ng Optimization of Offshore Saline Aquifer CO<sub>2</sub> Storage in Smeaheia Using Surrogate Reservoir Models Algorithms geological carbon storage surrogate reservoir model artificial intelligence deep learning optimization |
| title | Optimization of Offshore Saline Aquifer CO<sub>2</sub> Storage in Smeaheia Using Surrogate Reservoir Models |
| title_full | Optimization of Offshore Saline Aquifer CO<sub>2</sub> Storage in Smeaheia Using Surrogate Reservoir Models |
| title_fullStr | Optimization of Offshore Saline Aquifer CO<sub>2</sub> Storage in Smeaheia Using Surrogate Reservoir Models |
| title_full_unstemmed | Optimization of Offshore Saline Aquifer CO<sub>2</sub> Storage in Smeaheia Using Surrogate Reservoir Models |
| title_short | Optimization of Offshore Saline Aquifer CO<sub>2</sub> Storage in Smeaheia Using Surrogate Reservoir Models |
| title_sort | optimization of offshore saline aquifer co sub 2 sub storage in smeaheia using surrogate reservoir models |
| topic | geological carbon storage surrogate reservoir model artificial intelligence deep learning optimization |
| url | https://www.mdpi.com/1999-4893/17/10/452 |
| work_keys_str_mv | AT behzadamiri optimizationofoffshoresalineaquifercosub2substorageinsmeaheiausingsurrogatereservoirmodels AT ashkanjahanbanighahfarokhi optimizationofoffshoresalineaquifercosub2substorageinsmeaheiausingsurrogatereservoirmodels AT verarocca optimizationofoffshoresalineaquifercosub2substorageinsmeaheiausingsurrogatereservoirmodels AT cuthbertshangwuing optimizationofoffshoresalineaquifercosub2substorageinsmeaheiausingsurrogatereservoirmodels |