Artificial neural network prediction models for Montney shale gas production profile based on reservoir and fracture network parameters
Viet Nguyen-Le and
Hyundon Shin
Energy, 2022, vol. 244, issue PB
Abstract:
The prediction of shale gas production is necessary to evaluate the project's economical feasibility. Some studies suggested prediction models for predicting shale gas production. However, the model-based planar fracture assumption may not apply to a naturally fractured shale gas reservoir which induces a complex fracture network. This paper proposes three ANN architectures for predicting the peak production and Arps's hyperbolic decline parameters (Di and b) of a shale gas well in the Montney formation with an existing natural fracture system. A production profile can be reconstructed using the Arps' hyperbolic decline model and the predicted parameters. The ANN architectures were developed based on 370 simulation data of the reservoir, hydraulic fracture design parameters, and the fracture network properties, including fracture spacing and fracture conductivity, which remarkably affect shale gas production. The testing results, using another set of 92 simulation data, confirmed the high correlation between the input and objective functions with R2 > 0.86. Moreover, good agreement was observed between the measured and predicted cumulative gas production at one-, five-, ten-, fifteen-, and twenty-years of production with R2 > 0.94, and percentage errors were lower than 15.6%. This suggests that the shale gas production can be predicted efficiently and reliably using the Arps' hyperbolic model and the predicted parameters. The estimated production profiles can be used to continuously update the field development plans and calculate the project's NPV. Furthermore, the proposed method is applicable for predicting the production of newly produced reservoirs with limited production history.
Keywords: Shale gas; Prediction model; Artificial neural network; Peak production rate; Decline parameter; Fracture network (search for similar items in EconPapers)
Date: 2022
References: View complete reference list from CitEc
Citations: View citations in EconPapers (5)
Downloads: (external link)
http://www.sciencedirect.com/science/article/pii/S0360544222000536
Full text for ScienceDirect subscribers only
Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.
Export reference: BibTeX
RIS (EndNote, ProCite, RefMan)
HTML/Text
Persistent link: https://EconPapers.repec.org/RePEc:eee:energy:v:244:y:2022:i:pb:s0360544222000536
DOI: 10.1016/j.energy.2022.123150
Access Statistics for this article
Energy is currently edited by Henrik Lund and Mark J. Kaiser
More articles in Energy from Elsevier
Bibliographic data for series maintained by Catherine Liu ().