Digital twin-driven nonplanar fracture reconstruction and evaluation in tight reservoirsOA
In tight oil reservoir development,hydraulic fracture morphology critically controls post-fracturing flow behavior.However,conventional microseismic monitoring remains limited in real-time characterization of nonplanar fracture geometry.This study proposes a digital twin-driven approach for nonplanar fracture reconstruction and evaluation,enabling real-time fracture characterization in tight reservoirs.The results show that:(1)the reconstructed fracture morphology achieved a normalized relative geometric error of only 3.33%when experimentally validated against high-resolution laser scanning in a true triaxial hydraulic fracturing test,and demonstrated a computational speedup exceeding 700 times compared to the discrete element method under identical experimental conditions and specimen geometry;(2)the reconstructed fracture networks capture nonplanar propagation and fracture–natural fracture interactions,indicating that fracture complexity is jointly controlled by tortuosity and connectivity when the horizontal stress difference is below 5 MPa,but dominated by connectivity when it exceeds 5 MPa;(3)real-time fracture reconstruction enables the identification of potential inter-stage fracture communication,allowing engineers to adjust injection volume and stage spacing to mitigate interference and improve stimulation efficiency.These results demonstrate the proposed digital twin functions both as a monitoring tool and an integrated platform for real-time fracture reconstruction,evaluation,and operational optimization in tight reservoirs.
Botao Lin;Ji Lu;Jinyang Xie;Yan Jin;Han Meng
State Key Laboratory of Petroleum Resources and Engineering,China University of Petroleum(Beijing),Beijing 102249,China College of Artificial Intelligence,China University of Petroleum(Beijing),Beijing 102249,ChinaState Key Laboratory of Petroleum Resources and Engineering,China University of Petroleum(Beijing),Beijing 102249,China Department of Computer and Information Sciences,Centre for Research in Data Science,Universiti Teknologi PETRONAS,Seri Iskandar,Perak 32610,MalaysiaState Key Laboratory of Petroleum Resources and Engineering,China University of Petroleum(Beijing),Beijing 102249,ChinaState Key Laboratory of Petroleum Resources and Engineering,China University of Petroleum(Beijing),Beijing 102249,ChinaCollege of Artificial Intelligence,China University of Petroleum(Beijing),Beijing 102249,China
能源科技
Digital twinHydraulic fractureMicroseismic monitoringNonplanar fractureReal-time visualization
《Natural Gas Industry B》 2026 (3)
P.323-347,25
supported by the National Natural Science Foundation of China(Grant No.42277122).
评论