Identification of the formation temperature field of the southern Songliao Basin, China based on a deep belief network

Weifei Yang, Changlai Xiao, Zhihao Zhang and Xiujuan Liang

Renewable Energy, 2022, vol. 182, issue C, 32-42

Abstract: Since geothermal energy is clean and renewable, it has attracted the attention of governments and researchers globally. However, the cost of measurement of the middle-deep formation temperature of geothermal energy is relatively high. Therefore, usually only limited measured ground temperature data are available for regional-scale geothermal research. The limited measured ground temperature data hinders obtaining a relatively accurate characterization of the regional-scale formation temperature field. This study proposes an innovative method for identifying the formation temperature field based on a deep belief network (DBN). The method was applied to identify the formation temperature field in the southern Songliao Basin, Northeast China. The initial parameters of the DBN model were first optimized through pre-training, following which the optimal parameters were obtained through fine-tuning. This allowed the method to identify the complex relationship between the 20 identification factors and the measured temperature and finally to predict the formation temperature. The optimized model showed prediction deviations ranging from −5 °C and 5 °C and from −1 °C and 1 °C for 95% and 67% of the test dataset, respectively. This research provides a reference for an optimizing the accuracy at which the three-dimensional (3D) formation temperature can be identified based on the limited measured ground temperature data. This process is also suitable for other 2D or 3D field prediction studies with limited training data, such as for the prediction of formation porosity, thermal conductivity, and permeability.

Keywords: Deep belief network; Formation temperature; Geothermal gradient; Songliao basin (search for similar items in EconPapers)
Date: 2022
References: View references in EconPapers View complete reference list from CitEc
Citations: View citations in EconPapers (2)

Downloads: (external link)
http://www.sciencedirect.com/science/article/pii/S0960148121014658
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:renene:v:182:y:2022:i:c:p:32-42

DOI: 10.1016/j.renene.2021.09.127

Access Statistics for this article

Renewable Energy is currently edited by Soteris A. Kalogirou and Paul Christodoulides

More articles in Renewable Energy from Elsevier
Bibliographic data for series maintained by Catherine Liu ().