DeepONet-embedded physics-informed neural network for production prediction of multiscale shale matrix–fracture system

Physics of Fluids, Journal Year: 2025, Volume and Issue: 37(1)

Published: Jan. 1, 2025

As a rising method for reservoir-scale production analysis, machine learning (ML) models possess high computational efficiency with robust capability of nonlinear mapping. However, their accuracy and interpretability are commonly limited owing to the absence intrinsic physical mechanisms, solely by data fitting. This work proposes novel DeepONet-embedded physics-informed neural network (DE-PINN), which comprises forward connect matrix/fracture characteristics performance, sampling acquire location points within shale reservoirs. DeepONets constructed selected layers these networks output field variables in governing equations that include mass/momentum conservation coupled multiscale transport mechanisms. Through automatic differentiation method, solved obtained variables, residuals generated during solution integrated into loss function as constraints. Compared traditional data-driven models, DE-PINN exhibits better performance forecasting rate cumulative production, achieving mean absolute percentage error (MAPE) approximately 3% adjusted R2 values test set exceeding 0.98. model demonstrates advantage realizing superior predictive precision fewer samples under complex geological conditions

Language: Английский

---

A Quantitative Evaluation Method for Inter-well Fracture Communication Strength of Multifractured Horizontal Wells from Two-Phase Shale Gas Reservoirs: Field Cases Study

Published: April 21, 2025

Abstract The reduction in well spacing for multi-fractured horizontal wells shale gas reservoirs increases the risk of fracture communication. Fracture hits and induced communication between parent child can significantly reduce production well. current models are unable to quantify strength inter-well hydraulic simultaneously. This work develops a practical semi-analytical model quantification by analyzing two-phase data from parent-child system. We define hit region (FHR) enhanced fractured (EFHR) establish relationship fractures. simplifies system based on compartment concept, dividing flow into primary (PHF), (EFR), region. A set differential equations is solved using multi-region material balance with adaptive time-stepping obtain changes pressure saturation within drainage area calculate production. results show that undergoes transient decline after fracture. Once commences communication, rate exceeds natural decline. comparison numerical simulation validates accuracy new method. Subsequently, this method applied platform Sichuan China. History matching single-well explain parameters such as half-length, reservoir water saturation, matrix permeability, length permeability region, EFHR volume connected higher computational efficiency than simulation. quantitative evaluation communicating proposed. simultaneously fracture, provide guidance evaluating effectiveness fracturing effect.

Language: Английский

---

Modeling the relative permeability of gas and water flow in rock fractures with surface morphology parameters

Physics of Fluids, Journal Year: 2024, Volume and Issue: 36(12)

Published: Dec. 1, 2024

Rock fracture morphology influences two-phase fluid flow in rock fractures. However, the role of surface fractures remains to be fully quantified. In this study, an analytical model for gas and water relative permeability rough was developed, incorporating roughness factor, hydraulic tortuosity, tortuosity quantify morphology. The performance proposed evaluated through a comparison with empirical models experimental data. accurately characterized examined impact dynamics on fluids fractures, demonstrating that increased shifts equal-permeability point toward higher saturation lower permeability. To investigate influence permeability, extended express Reynolds number as function enabling detailed exploration evolution law number.

Language: Английский

---

A hydraulic-mechanical coupling model of dynamic waterflood-induced fractures in fractured tight reservoirs

Physics of Fluids, Journal Year: 2024, Volume and Issue: 36(12)

Published: Dec. 1, 2024

The propagation of waterflood-induced fractures (WIFs) occurs during prolonged water injection and is influenced by the distribution properties natural (NFs). Available numerical models rarely consider fracture activation rupture in an integrated manner, which makes it difficult to reflect complex morphology. In this paper, we propose a hydraulic-mechanical model with strain-dependent damage variables describe dynamic expansion characteristics WIFs. There are discrete filled NFs matrix non-equal-thickness joint elements, for derive constitutive equations calculate widths production. Damage calculated according maximum tensile stress criterion Mohr–Coulomb criterion. A comparison between coupled experimental results conducted demonstrate its validity. Finally, simulated analyzed four influencing factors pressure response evolution. study demonstrates that behavior area evolution highly sensitive rate, communication sequence, NF density, orientation. activation, cross, capture interactions WIFs complicate fracture-damage network enhance seepage efficiency. High rates promote crack tip propagation, while lower facilitate secondary at low pressure. For high density reservoirs, low-pressure fully activates NFs, aiding excessive induces simpler morphologies, making unstable more effective than continuous injection. This work guides appropriately induced improve absorption tight reservoirs.

Language: Английский

---