Akihiro Shima, Kazuya Ishitsuka, Weiren Lin, Elvar K. Bjarkason, Anna Suzuki
{"title":"Modeling unobserved geothermal structures using a physics-informed neural network with transfer learning of prior knowledge","authors":"Akihiro Shima, Kazuya Ishitsuka, Weiren Lin, Elvar K. Bjarkason, Anna Suzuki","doi":"10.1186/s40517-024-00312-7","DOIUrl":null,"url":null,"abstract":"<div><p>Deep learning has gained attention as a potentially powerful technique for modeling natural-state geothermal systems; however, its physical validity and prediction inaccuracy at extrapolation ranges are limiting. This study proposes the use of transfer learning in physics-informed neural networks to leverage prior expert knowledge at the target site and satisfy conservation laws for predicting natural-state quantities such as temperature, pressure, and permeability. A neural network pre-trained with multiple numerical datasets of natural-state geothermal systems was generated using numerical reservoir simulations based on uncertainties of the permeabilities, sizes, and locations of geological units. Observed well logs were then used for tuning by transfer learning of the network. Two synthetic datasets were examined using the proposed framework. Our results demonstrate that the use of transfer learning significantly improves the prediction accuracy in extrapolation regions with no observed wells.</p></div>","PeriodicalId":48643,"journal":{"name":"Geothermal Energy","volume":null,"pages":null},"PeriodicalIF":2.9000,"publicationDate":"2024-10-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://geothermal-energy-journal.springeropen.com/counter/pdf/10.1186/s40517-024-00312-7","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Geothermal Energy","FirstCategoryId":"89","ListUrlMain":"https://link.springer.com/article/10.1186/s40517-024-00312-7","RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"ENERGY & FUELS","Score":null,"Total":0}
引用次数: 0
Abstract
Deep learning has gained attention as a potentially powerful technique for modeling natural-state geothermal systems; however, its physical validity and prediction inaccuracy at extrapolation ranges are limiting. This study proposes the use of transfer learning in physics-informed neural networks to leverage prior expert knowledge at the target site and satisfy conservation laws for predicting natural-state quantities such as temperature, pressure, and permeability. A neural network pre-trained with multiple numerical datasets of natural-state geothermal systems was generated using numerical reservoir simulations based on uncertainties of the permeabilities, sizes, and locations of geological units. Observed well logs were then used for tuning by transfer learning of the network. Two synthetic datasets were examined using the proposed framework. Our results demonstrate that the use of transfer learning significantly improves the prediction accuracy in extrapolation regions with no observed wells.
Geothermal EnergyEarth and Planetary Sciences-Geotechnical Engineering and Engineering Geology
CiteScore
5.90
自引率
7.10%
发文量
25
审稿时长
8 weeks
期刊介绍:
Geothermal Energy is a peer-reviewed fully open access journal published under the SpringerOpen brand. It focuses on fundamental and applied research needed to deploy technologies for developing and integrating geothermal energy as one key element in the future energy portfolio. Contributions include geological, geophysical, and geochemical studies; exploration of geothermal fields; reservoir characterization and modeling; development of productivity-enhancing methods; and approaches to achieve robust and economic plant operation. Geothermal Energy serves to examine the interaction of individual system components while taking the whole process into account, from the development of the reservoir to the economic provision of geothermal energy.