EUVIMEDEUVIMED entdecken
Crossref

A Reliable Surrogate-Assisted Optimization Framework for CO2 Enhanced Oil Recovery and Sequestration in Unconventional Resources

Chaoxu Qin, Sen Wang, Xiaopeng Cao, Qihong Feng, Yulong Zhang

SPE Journal · 01.11.2026 · 1-22

DOI: 10.2118/236985-pa ↗

Kurz gefasst

Fachartikel in der Zeitschrift SPE Journal, online erschienen am 11.11.2026. Verfasst von Chaoxu Qin und 4 weiteren Autor:innen. Die Arbeit nennt 86 Literaturangaben.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

CO2 enhanced oil recovery (CO2-EOR) in unconventional resources has proven to be a promising method for carbon storage to mitigate severe global warming challenges. However, optimizing the operational parameters of CO2-EOR projects is time-consuming due to the high cost of running high-precision reservoir numerical simulations. A surrogate-assisted optimization algorithm is efficient in addressing such time-consuming problems, but ensuring the reliability of surrogate models with reasonable computational consumption is challenging. Therefore, this work proposes a new reliable ensemble surrogate-assisted differential evolution with a multistrategy restart (ESADE-MSR) framework to optimize operational parameters for CO2-EOR projects in unconventional resources. The optimization objective is to maximize the net present value (NPV) while considering the economic incentives created by tax credit policies to enhance CO2 sequestration. We adopt an ensemble surrogate model with three heterogeneous machine learning algorithms and introduce multistrategy sample updating with algorithm restart to improve optimization performance and surrogate reliability. A comparison between ESADE-MSR and conventional optimization methods is conducted on synthetic and real shale oil cases. The results demonstrate that under the same comparison conditions, ESADE-MSR achieves higher optimal NPV, with increases of 8.86% and 5.43%, and improves efficiency by 66.48% and 41.02%, respectively. This work provides a powerful tool for designing reasonable operational parameters for CO2-EOR and sequestration in unconventional resources and has implications for solving similar time-consuming optimization problems in other energy fields.

Zitieren

APA

Qin, C., Wang, S., Cao, X., Feng, Q., & Zhang, Y. (2026). A Reliable Surrogate-Assisted Optimization Framework for CO2 Enhanced Oil Recovery and Sequestration in Unconventional Resources. SPE Journal, 1-22. https://doi.org/10.2118/236985-pa

Vancouver

Qin C, Wang S, Cao X, Feng Q, Zhang Y. A Reliable Surrogate-Assisted Optimization Framework for CO2 Enhanced Oil Recovery and Sequestration in Unconventional Resources. SPE Journal. 2026:1-22. doi:10.2118/236985-pa

BibTeX
@article{Qin2026,
  author = {Qin, Chaoxu and Wang, Sen and Cao, Xiaopeng and Feng, Qihong and Zhang, Yulong},
  title = {A Reliable Surrogate-Assisted Optimization Framework for CO2 Enhanced Oil Recovery and Sequestration in Unconventional Resources},
  journal = {SPE Journal},
  year = {2026},
  pages = {1--22},
  doi = {10.2118/236985-pa}
}
Kostenlos starten →