Towards better shale gas production forecasting using transfer learning

• Published in: Upstream Oil and Gas Technology Vol. 9; p. 100072
• Main Authors: Alolayan, Omar S., Raymond, Samuel J., Montgomery, Justin B., Williams, John R.
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01.09.2022
• Subjects: County-specific models; Decline curve analysis; Machine learning; Shale gas forecasting; Transfer learning; Decline curve analysis; Shale gas forecasting; County-specific models; Transfer learning; Machine learning
• ISSN: 2666-2604; 2666-2604
• DOI: 10.1016/j.upstre.2022.100072

Deep neural networks can generate more accurate shale gas production forecasts in counties with a limited number of sample wells by utilizing transfer learning. This paper provides a way of transferring the knowledge gained from other deep neural network models trained on adjacent counties into the county of interest. The paper uses data from more than 6000 shale gas wells across 17 counties from Texas Barnett and Pennsylvania Marcellus shale formations to test the capabilities of transfer learning. The results reduce the forecasting error between 11% and 47% compared to the widely used Arps decline curve model.