Production simulation is an important method to evaluate the stimulation effect of refracturing. Therefore, a production simulation model based on coupled fluid flow and geomechanics in triple continuum including kerogen, an inorganic matrix, and a fracture network is proposed considering the multiscale flow characteristics of shale gas, the induced stress of fracture opening, and the pore elastic effect. The complex transport mechanisms due to multiple physics, including gas adsorption/desorption, slip flow, Knudsen diffusion, surface diffusion, stress sensitivity, and adsorption layer are fully considered in this model. The apparent permeability is used to describe the multiple physics occurring in the matrix. The model is validated using actual production data of a horizontal shale gas well and applied to predict the production and production increase percentage (PIP) after refracturing. A sensitivity analysis is performed to study the effects of the refracturing pattern, fracture conductivity, width of stimulated reservoir volume (SRV), SRV length of new and initial fractures, and refracturing time on production and the PIP. In addition, the effects of multiple physics on the matrix permeability and production, and the geomechanical effects of matrix and fracture on production are also studied. The research shows that the refracturing design parameters have an important influence on the PIP. The geomechanical effect is an important cause of production loss, while slippage and diffusion effects in matrix can offset the production loss.

The productivity evaluation of CBM wells can yield significant insights into exploring the patterns of CBM production, predicting the effectiveness of the CBM well and reservoir stimulation, optimizing the gas reservoir development program, and developing a reasonable production system, for the purpose of facilitating efficient development of CBM. In particular, to accurately evaluate CBM productivity, we should establish the corresponding mathematical model of fluid flow through porous media and productivity evaluation model based on a clear understanding of CBM occurrence states and mechanisms of its flow through porous media. After considering the effects of slip flow, Knudsen diffusion, surface diffusion, stress sensitivity, and matrix shrinkage on fluid mass transfer, we have put forward a triple-porosity and dual-permeability mathematical model to predict CBM productivity that incorporates matrix gas desorption, complex flow in matrix pores, and gas–water two-phase flow in a cleat system. In combination with reservoir characteristic parameters, a case study of Ma-26 well in the Mabidong block in the south of the Qinshui Basin, we carried out a numerical simulation of the productivity of a fractured CBM well and analyzed the effects (on production performance) of occurrence states, cleat system permeability, complex flow regimes in the matrix, Langmuir pressure, and Langmuir volume on production are provided. The results show that 1) in the initial drainage and production stage of CBM wells, both free gas and adsorbed gas are produced simultaneously, while adsorbed gas dominates the production in later stages; 2) the peak output and cumulative output of CBM wells increase significantly with the rise in cleat system permeability; 3) the increase of Langmuir pressure, volume, and matrix porosity are conducive to the increase of CBM production. The research has considerable reference value for work on mechanisms of CBM flow in porous media and post fracturing productivity evaluation of CBM and also provides a theoretical basis for fieldwork in CBM development.

Stimulated reservoir volume (SRV) is a key parameter affecting the post-fracturing production of ultra-low permeability reservoirs. Accurate prediction of the SRV shape and size is of vital significance for pre-fracturing stimulation design and post-fracturing effect evaluation. In this paper, an equivalent mathematical model for simulating two-dimensional SRV growth is proposed by integrating the fracture width equation, tensor permeability conversion, cubic law and calculation method of fracture porosity with mass conservation equation of injected fluid in natural fracture system. The finite difference method is used to derive the numerical discrete equation, and the fluid pressure and average fracture width in each grid block are obtained by solving the coupled fluid conservation equation and fracture width equation. The model is validated by comparing the results obtained from this work with others from literature. A parametric sensitivity study was performed to analyze the effects of some factors incorporating approaching angle, natural fracture spacing, horizontal principal stress difference, pumping rate and fracturing fluid viscosity on the SRV (including size, aspect ratio, and arithmetic mean fracture width in SRV region). The results show that SRV increases with the decrease of approaching angle, principal stress difference, injection rate, viscosity of fracturing fluid and fracture density. The width of branch fracture decreases with the increasing approaching angle and principal stress difference but increases with the increasing pumping rate and fluid viscosity.

Influenced by natural fractures, fracture networks may be created when hydraulic fractures propagate in naturally fractured reservoirs. Simulating fracture network propagation is crucial to fracture design. In this paper, a two-dimensional mathematical model for the direct simulation of fracture network propagation that considers fluid filtration and reservoir fluid flow is developed, and a simulation method is proposed based on the coupled flow idea of natural fractures and matrices in dual-porosity media. The model uses dynamic coordinates to represent the extended hydraulic fracture network and static coordinates to describe natural fractures that have not been activated. The activation of natural fractures by a hydraulic fracture is determined according to the hydraulic fracture propagation scenario after an intersection with natural fractures. The geometric shape and size of the network fracture are obtained by changing the size of the simulated area and solving the mass balance, fracture width, and reservoir fluid flow equations numerically in a coupled manner. After validating the new model, a parameter sensitivity analysis is conducted to study the effects of injected fluid volume, fracture height, fracture spacing, elastic modulus, horizontal principal stress difference, pumping rate, and fracturing fluid viscosity on the fracture network shape, fracture network size, average fracture width, and fracture-reservoir contact area. The numerical results indicate that the model can be directly used to simulate the propagation of fracture networks. The volume of injected fluid is the most critical factor affecting the fracture network size. Increasing the fluid viscosity and injection rate can increase the width of the fracture network and the average width of a secondary fracture. However, the total contact area of the reservoir and fracture will decrease, and the reservoir-fracture contact area and fracture network size will increase with a decrease in the horizontal stress difference and increase in elastic modulus. Decreasing a fracture's height can significantly increase the size of the fracture network and reservoir-fracture contact area.