EUVIMEDEUVIMED entdecken
Crossref

Inversion Framework of Reservoir Parameters Based on Deep Autoregressive Surrogate and Continual Learning Strategy

Kai Zhang, Wenhao Fu, Jinding Zhang, Wensheng Zhou, Chen Liu, Piyang Liu, Liming Zhang, Xia Yan, Yongfei Yang, Hai Sun, Jun Yao

SPE Journal · 11.10.2023 · 28 (05) · 2223-2239

DOI: 10.2118/215821-pa ↗

Kurz gefasst

Fachartikel in der Zeitschrift SPE Journal, online erschienen am 02.06.2023. Verfasst von Kai Zhang und 10 weiteren Autor:innen. Die Arbeit wurde bisher 14-mal zitiert (Crossref) und nennt 58 Literaturangaben.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

History matching is a crucial process that enables the calibration of uncertain parameters of the numerical model to obtain an acceptable match between simulated and observed historical data. However, the implementation of the history-matching algorithm is usually based on iteration, which is a computationally expensive process due to the numerous runs of the simulation. To address this challenge, we propose a surrogate model for simulation based on an autoregressive model combined with a convolutional gated recurrent unit (ConvGRU). The proposed ConvGRU-based autoregressive neural network (ConvGRU-AR-Net) can accurately predict state maps (such as saturation maps) based on spatial and vector data (such as permeability and relative permeability, respectively) in an end-to-end fashion. Furthermore, history matching must be performed multiple times throughout the production cycle of the reservoir to fit the most recent production observations, making continual learning crucial. To enable the surrogate model to quickly learn recent data by transferring experience from previous tasks, an ensemble-based continual learning strategy is used. Together with the proposed neural network–based surrogate model, the randomized maximum likelihood (RML) is used to calibrate uncertain parameters. The proposed method is evaluated using 2D and 3D reservoir models. For both cases, the surrogate inversion framework successfully achieves a reasonable posterior distribution of reservoir parameters and provides a reliable assessment of the reservoir’s behaviors.

Zitieren

APA

Zhang, K., Fu, W., Zhang, J., Zhou, W., Liu, C., Liu, P., Zhang, L., Yan, X., Yang, Y., Sun, H., & Yao, J. (2023). Inversion Framework of Reservoir Parameters Based on Deep Autoregressive Surrogate and Continual Learning Strategy. SPE Journal, 28(05), 2223-2239. https://doi.org/10.2118/215821-pa

Vancouver

Zhang K, Fu W, Zhang J, Zhou W, Liu C, Liu P, et al. Inversion Framework of Reservoir Parameters Based on Deep Autoregressive Surrogate and Continual Learning Strategy. SPE Journal. 2023;28(05):2223-2239. doi:10.2118/215821-pa

BibTeX
@article{Zhang2023,
  author = {Zhang, Kai and Fu, Wenhao and Zhang, Jinding and Zhou, Wensheng and Liu, Chen and Liu, Piyang and Zhang, Liming and Yan, Xia and Yang, Yongfei and Sun, Hai and Yao, Jun},
  title = {Inversion Framework of Reservoir Parameters Based on Deep Autoregressive Surrogate and Continual Learning Strategy},
  journal = {SPE Journal},
  year = {2023},
  volume = {28},
  number = {05},
  pages = {2223--2239},
  doi = {10.2118/215821-pa}
}
Kostenlos starten →