EUVIMEDEUVIMED entdecken
VergleichsstudieOpen Access · CC BY 4.0

A comparative study of MLP and LSTM neural networks for shale gas production prediction based on numerical simulation data

Xiaoou Fei, Man Ye, Zhou Du, Haibin Miao iD

PLOS One · 13.11.2025 · 20 (11) · e0336782

DOI: 10.1371/journal.pone.0336782 ↗

Kurz gefasst

Vergleichsstudie in der Zeitschrift PLOS One, online erschienen am 13.11.2025. Verfasst von Xiaoou Fei und 3 weiteren Autor:innen. Beteiligte Einrichtung: School of Green Mining and Resource Engineering, Liaoning Petrochemical University, Fushun, Liaoning, China. Hauptthemen laut MEDLINE: Neural Networks, Computer, Natural Gas, Deep Learning. Die Arbeit wurde bisher 4-mal zitiert (Crossref) und nennt 27 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Liaoning Provincial Department of Education General Project, National Key Research and Development Program, Liaoning Province Applied Basic Research Program.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Accurate prediction of shale gas production is essential for optimizing reservoir development and improving production efficiency. In this study, a numerical simulation model was first developed to systematically calculate daily shale gas production under various engineering parameter combinations, thereby establishing a comprehensive production prediction database. Two types of deep learning models—multi-layer perceptron (MLP) and long short-term memory (LSTM) neural networks—were then constructed to predict daily shale gas production. Comparisons with actual production data for three representative scenarios revealed that the MLP model achieved relative errors of 2.43%, 6.36%, and 4.16%, while the LSTM model achieved superior accuracy with relative errors of 0.42%, 1.1%, and 0.98%. The LSTM network’s gating mechanisms effectively captured the long-term dependencies in shale gas production data, making it more suitable for complex multi-scale dynamic modeling compared to the feedforward MLP. These results demonstrate the excellent generalization capability and engineering applicability of deep learning techniques, particularly LSTM networks, for enhancing shale gas production forecasting and supporting the efficient development of unconventional gas reservoirs.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Neural Networks, ComputerNatural GasDeep LearningComputer Simulation

Substanzen

Natural Gas

Autor:innen und Institutionen

  • School of Green Mining and Resource Engineering, Liaoning Petrochemical University, Fushun, Liaoning, China.
  • School of Mechanics and Engineering, Liaoning Technical University, Fuxin, Liaoning, China.
  • China Coal Technology and Engineering Group Shenyang Research Institute, Fushun, Liaoning, China.
  • State Key Laboratory of Disaster Prevention and Ecology Protection in Open-pit Coal Mines, Fushun, Liaoning, China.

ORCID-Profile: Haibin Miao

Förderung

  • Liaoning Provincial Department of Education General Project
  • National Key Research and Development Program
  • Liaoning Province Applied Basic Research Program

Zitieren

APA

Fei, X., Ye, M., Du, Z., & Miao, H. (2025). A comparative study of MLP and LSTM neural networks for shale gas production prediction based on numerical simulation data. PLOS One, 20(11), e0336782. https://doi.org/10.1371/journal.pone.0336782

Vancouver

Fei X, Ye M, Du Z, Miao H. A comparative study of MLP and LSTM neural networks for shale gas production prediction based on numerical simulation data. PLOS One. 2025;20(11):e0336782. doi:10.1371/journal.pone.0336782

BibTeX
@article{Fei2025,
  author = {Fei, Xiaoou and Ye, Man and Du, Zhou and Miao, Haibin},
  title = {A comparative study of MLP and LSTM neural networks for shale gas production prediction based on numerical simulation data},
  journal = {PLOS One},
  year = {2025},
  volume = {20},
  number = {11},
  pages = {e0336782},
  doi = {10.1371/journal.pone.0336782}
}
Kostenlos starten →