TY - JOUR

T1 - A physics-boosted transfer learning framework for fracturing pressure prediction with scarce data

AU - Hou, Lei

AU - Luo, Jiangfeng

AU - Dontsov, Egor

AU - Zhang, Zhengxin

AU - Valov, Alexander

AU - Zhang, Fengshou

AU - Bian, Xiaobing

AU - Fu, Liang

N1 - This research is funded by the National Key Research and Development Project (No. 2023YFE0110900), National Natural Science Foundation of China under the grant 42377138, and the Open Fund of State Key Laboratory of Shale Oil and Gas Enrichment Mechanisms and Efficient Development.

PY - 2026/2

Y1 - 2026/2

N2 - Accurately predicting fracturing pressure is critical for optimizing the safety and efficiency of hydraulic fracturing operations, particularly in newly developed blocks where data scarcity poses significant challenges. Traditional machine learning methods require large, high-quality datasets to train algorithms. To address these limitations, this study presents physics-boosted transfer learning frameworks designed to enhance fracturing pressure prediction in data-scarce scenarios. By integrating a gated recurrent unit (GRU) deep learning model with physical modeling principles, three transfer learning frameworks were developed and evaluated, including a pure data-driven framework, a hybrid-modelling framework, and a physics-informed framework. Field data from only three shale gas wells were utilized to train the GRU algorithm – simulating real-field data-scarcity scenarios. Fine-tuning technologies are optimized based on the pure data-driven framework. The physics-informed framework demonstrated superior performance, achieving root mean square errors (RMSE) as low as 2–3 MPa, significantly outperforming both the pure data-driven and hybrid frameworks in terms of accuracy, stability, and adaptability. By bridging the gap between data-driven methods and physical modeling, this new framework offers a robust solution, for improving operational safety and cost-effectiveness in hydraulic fracturing, particularly under data-scarce conditions.

AB - Accurately predicting fracturing pressure is critical for optimizing the safety and efficiency of hydraulic fracturing operations, particularly in newly developed blocks where data scarcity poses significant challenges. Traditional machine learning methods require large, high-quality datasets to train algorithms. To address these limitations, this study presents physics-boosted transfer learning frameworks designed to enhance fracturing pressure prediction in data-scarce scenarios. By integrating a gated recurrent unit (GRU) deep learning model with physical modeling principles, three transfer learning frameworks were developed and evaluated, including a pure data-driven framework, a hybrid-modelling framework, and a physics-informed framework. Field data from only three shale gas wells were utilized to train the GRU algorithm – simulating real-field data-scarcity scenarios. Fine-tuning technologies are optimized based on the pure data-driven framework. The physics-informed framework demonstrated superior performance, achieving root mean square errors (RMSE) as low as 2–3 MPa, significantly outperforming both the pure data-driven and hybrid frameworks in terms of accuracy, stability, and adaptability. By bridging the gap between data-driven methods and physical modeling, this new framework offers a robust solution, for improving operational safety and cost-effectiveness in hydraulic fracturing, particularly under data-scarce conditions.

KW - Hydraulic fracturing

KW - Pressure prediction

KW - Scarce data

KW - Shale gas

KW - Transfer learning

UR - https://www.scopus.com/pages/publications/105014804601

UR - https://www.mendeley.com/catalogue/351f2932-d1d9-321c-863a-f71d639701eb/

U2 - 10.1016/j.geoen.2025.214176

DO - 10.1016/j.geoen.2025.214176

M3 - Article

VL - 257

JO - Geoenergy Science and Engineering

JF - Geoenergy Science and Engineering

SN - 2949-8910

M1 - 214176

ER -