EUVIMEDEUVIMED entdecken
Crossref

Active Magnetic Ranging for Abandoned Well Plugging in Underground Hydrogen Storage: Fast Forward Prediction of Active Magnetic Field Around the Target Well

Zhe Liu, Jincheng Huang, Binbin Diao, Deli Gao, Huapeng Wu, Guanxiang Ding

SPE Journal · 11.11.2026 · 31 (11) · 7399-7420

DOI: 10.2118/236619-pa ↗

Kurz gefasst

Fachartikel in der Zeitschrift SPE Journal, online erschienen am 03.09.2026. Verfasst von Zhe Liu und 5 weiteren Autor:innen. Die Arbeit nennt 28 Literaturangaben.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Active magnetic ranging (AMR) technology shows strong potential for abandoned well sealing in underground hydrogen storage. However, coupled effects of formation electrical properties, target well structure, and operating parameters can significantly affect ranging accuracy. Based on the AMR principles, this study designs an adjacent well casing excitation and magnetic detection device. Physical experiments and numerical simulations are combined to reveal the distribution of magnetic field strength around the target casing. A sample database for the defined local AMR ranging domain is built using operating parameters, formation parameters, and target well structural parameters. A rapid surrogate model for forward prediction is then developed using a Stacking ensemble framework and a Gaussian-weighted mean squared error loss function (GWMSE). The model rapidly predicts the axial distribution of magnetic field strength around the target well. Predicted magnetic field strength curves support probe-position optimization and provide forward information for subsequent adjacent well ranging analysis. Among the compared models, Stacking-LR (linear regression) shows the best overall performance, with R2, mean absolute error (MAE), and root mean squared error (RMSE) reaching 0.928, 0.1986, and 0.5283, respectively. Compared with conventional numerical simulation, the proposed surrogate model reduces the computational cost of repeated forward predictions while maintaining stable prediction performance within the defined parameter domain and geometric assumptions. The results indicate that the proposed framework can provide rapid magnetic field forward prediction for probe-position optimization and subsequent adjacent well ranging analysis during the local fine-ranging stage of abandoned well sealing in underground hydrogen storage.

Zitieren

APA

Liu, Z., Huang, J., Diao, B., Gao, D., Wu, H., & Ding, G. (2026). Active Magnetic Ranging for Abandoned Well Plugging in Underground Hydrogen Storage: Fast Forward Prediction of Active Magnetic Field Around the Target Well. SPE Journal, 31(11), 7399-7420. https://doi.org/10.2118/236619-pa

Vancouver

Liu Z, Huang J, Diao B, Gao D, Wu H, Ding G. Active Magnetic Ranging for Abandoned Well Plugging in Underground Hydrogen Storage: Fast Forward Prediction of Active Magnetic Field Around the Target Well. SPE Journal. 2026;31(11):7399-7420. doi:10.2118/236619-pa

BibTeX
@article{Liu2026,
  author = {Liu, Zhe and Huang, Jincheng and Diao, Binbin and Gao, Deli and Wu, Huapeng and Ding, Guanxiang},
  title = {Active Magnetic Ranging for Abandoned Well Plugging in Underground Hydrogen Storage: Fast Forward Prediction of Active Magnetic Field Around the Target Well},
  journal = {SPE Journal},
  year = {2026},
  volume = {31},
  number = {11},
  pages = {7399--7420},
  doi = {10.2118/236619-pa}
}
Kostenlos starten →