Summary
Despite the progress in renewable energies, oil and gas remain the primary source of energy. Recovery from hydrocarbon reservoirs is subjected to three steps: primary, secondary and tertiary. The primary step results from the intrinsic reservoir energy; the second stage usually consists of injection of water or gas to support the pressure; and the third stage is the process of extracting the oil that cannot be recovered during the previous stages, by injecting miscible gas, thermal and chemicals. To assess the performance of implemented methods during the recovery steps, reservoir simulations are usually performed. However, these traditional simulations are known to be time-consuming, and significant number of runs is required to achieve optimal results. This project will use a combination of advanced methods including optimization, statistics and data-driven techniques, to develop a novel strategy for establishing user-friendly smart proxy models which aim at reducing significantly the run-time in reservoir simulation tasks. The project will be performed at four levels: the physical and numerical aspects of the recovery methods, sampling strategies to select runs for the proxy, learning techniques to build the proxy, and their application for optimizing recovery plans. The project has ultimate multidisciplinary aspects, including reservoir engineering, data science and environment (as CO2 storage is included in the project). The project will be carried out by the experienced researcher who worked during his PhD on the application of data-driven techniques for resolving petroleum engineering problems. The experienced researcher will collaborate with supervisors with a strong background in reservoir simulation and optimization. The transfer of knowledge from the project will have a twofold benefit, to the host institution and to the researcher. Expected results have the potential to improve the knowledge about simulation calculability using new robust approaches.
Unfold all
/
Fold all
More information & hyperlinks
| Web resources: | https://cordis.europa.eu/project/id/895406 |
| Start date: | 15-09-2021 |
| End date: | 14-09-2023 |
| Total budget - Public funding: | 202 158,72 Euro - 202 158,00 Euro |
Cordis data
Original description
Despite the progress in renewable energies, oil and gas remain the primary source of energy. Recovery from hydrocarbon reservoirs is subjected to three steps: primary, secondary and tertiary. The primary step results from the intrinsic reservoir energy; the second stage usually consists of injection of water or gas to support the pressure; and the third stage is the process of extracting the oil that cannot be recovered during the previous stages, by injecting miscible gas, thermal and chemicals. To assess the performance of implemented methods during the recovery steps, reservoir simulations are usually performed. However, these traditional simulations are known to be time-consuming, and significant number of runs is required to achieve optimal results. This project will use a combination of advanced methods including optimization, statistics and data-driven techniques, to develop a novel strategy for establishing user-friendly smart proxy models which aim at reducing significantly the run-time in reservoir simulation tasks. The project will be performed at four levels: the physical and numerical aspects of the recovery methods, sampling strategies to select runs for the proxy, learning techniques to build the proxy, and their application for optimizing recovery plans. The project has ultimate multidisciplinary aspects, including reservoir engineering, data science and environment (as CO2 storage is included in the project). The project will be carried out by the experienced researcher who worked during his PhD on the application of data-driven techniques for resolving petroleum engineering problems. The experienced researcher will collaborate with supervisors with a strong background in reservoir simulation and optimization. The transfer of knowledge from the project will have a twofold benefit, to the host institution and to the researcher. Expected results have the potential to improve the knowledge about simulation calculability using new robust approaches.Status
TERMINATEDCall topic
MSCA-IF-2019Update Date
28-04-2024
Images
No images available.
Geographical location(s)
Search
