Production prediction plays an important role in decision making, development planning, and economic evaluation during the exploration and development period. However, applying traditional methods for production forecasting of newly developed wells in the conglomerate reservoir is restricted by limited historical data, complex fracture propagation, and frequent operational changes. This study proposed a Gated Recurrent Unit (GRU) neural network-based model to achieve batch production forecasting in M conglomerate reservoir of China, which tackles the limitations of traditional decline curve analysis and conventional time-series prediction methods. The model is trained by four features of production rate, tubing pressure (TP), choke size (CS), and shut-in period (SI) from 70 multistage hydraulic fractured horizontal wells. Firstly, a comprehensive data preprocessing is implemented, including excluding unfit wells, data screening, feature selection, partitioning data set, z-score normalization, and format conversion. Then, the four-feature model is compared with the model considering production only, and it is found that with frequent oilfield operations changes, the four-feature model could accurately capture the complex variance pattern of production rate. Further, Random Forest (RF) is employed to optimize the prediction results of GRU. For a fair evaluation, the performance of the proposed model is compared with that of simple Recurrent Neural Network (RNN) and Long Short-Term Memory (LSTM) neural network. The results show that the proposed approach outperforms the others in prediction accuracy and generalization ability. It is worth mentioning that under the guidance of continuous learning, the GRU model can be updated as soon as more wells become available.

The water leak-off during hydraulic fracturing in shale gas reservoirs is a complicated transport behavior involving thermal (T), hydrodynamic (H), mechanical (M) and chemical (C) processes. Although many leak-off models have been published, none of the models fully coupled the transient fluid flow modeling with heat transfer, chemical-potential equilibrium and natural-fracture dilation phenomena. In this paper, a coupled thermo-hydro-mechanical-chemical (THMC) model based on non-equilibrium thermodynamics, hydrodynamics, thermo-poroelastic rock mechanics, and non-isothermal chemical-potential equations is presented to simulate the water leak-off process in shale gas reservoirs. The THMC model takes into account a triple-porosity medium, which includes hydraulic fractures, natural fractures and shale matrix. The leak-off simulation with the THMC model involves all the important processes in this triple-porosity medium, including: (1) water transport driven by hydraulic, capillary, chemical and thermal osmotic convections; (2) gas transport induced by both hydraulic pressure driven convection and adsorption; (3) heat transport driven by thermal convection and conduction; and (4) natural-fracture dilation considered as a thermo-poroelastic rock deformation. The fluid and heat transport, coupled with rock deformation, are described by a set of partial differential equations resulting from the conservation of mass, momentum, and energy. The semi-implicit finite-difference algorithm is proposed to solve these equations. The evolution of pressure, temperature, saturation and salinity profiles of hydraulic fractures, natural fractures and matrix is calculated, revealing the multi-field coupled water leak-off process in shale gas reservoirs. The influences of hydraulic pressure, natural-fracture dilation, chemical osmosis and thermal osmosis on water leak-off are investigated. Results from this study are expected to provide a better understanding of the predominant leak-off mechanisms for slickwater fracturing-fluids in hydraulically fractured shale gas reservoirs.

Water‐rock interaction cannot be ignored for shale reservoirs with high‐salinity formation brine and complex rock composition, and stimulated by massive slick‐water fracturing treatment. However, there have been few studies on the flowback model fully coupled with different effects of water‐rock interaction. This paper presents the development of a coupled hydro‐chemical‐mechanical model for modeling water‐rock interaction in fractured shale during the post‐fracturing flowback period. The model considers distinguishing water‐rock interaction phenomena, that is, mineral dissolution, clay swelling and chemical osmosis, and accounts for multi‐phase flow in a fractured shale reservoir. The coupling and solution method and a numerical simulator were developed. Numerical simulation indicates that the swelling volume of clay minerals occupies the pores and leads to a decline in matrix porosity, while mineral dissolution increases both the matrix porosity and the solute concentration in the aqueous phase in matrix pores. Clay swelling mainly affects the shape of the porosity ratio profiles. The effect of mineral dissolution becomes increasingly stronger as flowback progresses. Mineral dissolution mainly affects the relative positions of the porosity ratio curves with the progress of flowback. The water‐rock interaction coupled flowback modeling and the numerical simulation results in this study quantify the effects of chemical osmosis, clay swelling and mineral dissolution. Results from this study provide new insights into the mechanisms of fracturing‐fluid flowback and value to flowback transient analysis.

Post-fracturing well shut-in is traditionally due to the elastic closure of hydraulic fractures and proppant compaction. However, for shale gas wells, the extension of shut-in time may improve the post-fracturing gas production due to formation energy supplements by fracturing-fluid imbibition. This paper presents a methodology using numerical simulation to simulate the hydrodynamic equilibrium phenomenon of a hydraulically fractured shale gas reservoir, including matrix imbibition and fracture network crossflow, and further optimize the post-fracturing shut-in time. A mathematical model, which can describe the fracturing-fluid hydrodynamic transport during the shut-in process, and consider the distinguishing imbibition characteristics of a hydraulically fractured shale reservoir, i.e., hydraulic pressure, capillarity and chemical osmosis, is developed. The key concept, i.e., hydrodynamic equilibrium time, for optimizing the post-fracturing shut-in schedule, is proposed. The fracturing-fluid crossflow and imbibition profiles are simulated, which indicate the water discharging and sucking equilibrium process in the coupled fracture–matrix system. Based on the simulation, the hydrodynamic equilibrium time is calculated. The influences of hydraulic pressure difference, capillarity and chemical osmosis on imbibition volume, and hydrodynamic equilibrium time are also investigated. Finally, the optimal shut-in time is determined if the gas production rate is pursued and the fracturing-fluid loss is allowable. The proposed simulation method for determining the optimal shut-in time is meaningful to the post-fracturing shut-in schedule.