收藏切换
An intelligent prediction method for deep fluid storage potential based on machine learning
收藏切换
PDF
Zhan YANG1, 2, Xiang LI1, 2, Xiao ZHANG1, 2, Sheng TAO1, 2, Song DU1, 2, *
Science & Technology Review | 2026, 44(6) : 68 - 75
Less
收藏切换
Science & Technology Review | 2026, 44(6): 68-75
Exclusive
An intelligent prediction method for deep fluid storage potential based on machine learning
Full
Zhan YANG1, 2, Xiang LI1, 2, Xiao ZHANG1, 2, Sheng TAO1, 2, Song DU1, 2, *
Affiliations
  • 1General Prospecting and Research Institute of China National Administration of Coal Geology, Beijing 100039, China
  • 2Key Laboratory of Transparent Mine Geology and Digital Twin Technology, National Mine Safety Administration, Beijing 100039, China
Published: 2026-03-28 doi: 10.3981/j.issn.1000-7857.2025.12.00122
Outline
收藏切换

An intelligent prediction framework for deep fluid storage potential is proposed by integrating operational monitoring data with reservoir structural features. By introducing machine learning and time−series feature–driven modeling strategies, the proposed workflow enables automated identification of injection pressure evolution patterns and accurate prediction of storage−stage responses. The method combines long−term high−frequency pressure–flow monitoring data with pore–fracture structural information derived from core−scale analysis, and establishes a stage−labeled, phase−wise prediction mechanism to improve model robustness through multi−model comparison and cross−validation. The framework was validated in a deep injection and storage engineering scenario involving highly saline fluids, using an 18−month field dataset with a cumulative injection volume exceeding 1.5 million tonnes. Results demonstrate that the proposed intelligent workflow can effectively distinguish key operational stages, including breakthrough and filling phases, and significantly enhance pressure prediction accuracy, with the best model achieving a MAPE as low as 0.6. The method reduces dependence on complex mechanistic models and large labeled datasets, and shows strong generalization across different operational periods and stage conditions. Owing to its scalability and transferability, the proposed approach provides technical support for intelligent assessment and safe operation of various deep fluid injection and storage projects.

deep geological storage  /  deep geological fluid  /  pore pressure  /  machine learning  /  deep learning
Zhan YANG, Xiang LI, Xiao ZHANG, Sheng TAO, Song DU. An intelligent prediction method for deep fluid storage potential based on machine learning[J]. Science & Technology Review, 2026 , 44 (6) : 68 -75 . DOI: 10.3981/j.issn.1000-7857.2025.12.00122
Year 2026 volume 44 Issue 6
PDF
157
83
Cite this Article
BibTeX
Article Info
doi: 10.3981/j.issn.1000-7857.2025.12.00122
  • Receive Date:2025-11-21
  • Online Date:2026-04-16
  • Published:2026-03-28
Article Data
Affiliations
History
  • Received:2025-11-21
  • Revised:2026-02-10
Funding
Affiliations
    1General Prospecting and Research Institute of China National Administration of Coal Geology, Beijing 100039, China
    2Key Laboratory of Transparent Mine Geology and Digital Twin Technology, National Mine Safety Administration, Beijing 100039, China
References
Share
https://castjournals.cast.org.cn/joweb/kjdb/EN/10.3981/j.issn.1000-7857.2025.12.00122
Share to
QR

Scan QR to access full text

Cite this article
BibTeX
Citations
表12种不同金属材料的力学参数

Family
属数
Number of
genus
种数
Number of
species
占总种数比例
Percentage of
total species (%)

Genus
种数
Number of
species
占总种数比例
Percentage of total
species (%)
鹅膏菌科Amanitaceae 2 11 5.26 鹅膏菌属 Amanita 10 4.78
小菇科 Mycenaceae 2 12 5.74 丝盖伞属 Inocybe 5 2.39
多孔菌科 Polyporaceae 8 14 6.70 蜡蘑属 Laccaria 5 2.39
红菇科 Russulaceae 3 23 11.00 小皮伞属 Marasmius 6 2.87
小菇属 Mycena 11 5.26
光柄菇属 Pluteus 5 2.39
红菇属 Russula 17 8.13
栓菌属 Trametes 5 2.39
关闭全屏
  • BibTeX
  • EndNote
  • RefWorks
  • TxT