Understanding Shale Gas Flow Behavior Using Numerical Simulation

Abstract

Summary Shale gas has changed the energy equation around the world, and its impact has been especially profound in the United States. It is now generally agreed that the fabric of shale systems comprises primarily organic matter, inorganic material, and natural fractures. However, the underlying flow mechanisms through these multiporosity and multipermeability systems are poorly understood. For instance, debate still exists about the predominant transport mechanism (diffusion, convection, and desorption), as well as the flow interactions between organic matter, inorganic matter, and fractures. Furthermore, balancing the computational burden of precisely modeling the gas transport through the pores vs. running full reservoir scale simulation is also contested. To that end, commercial reservoir simulators are developing new shale gas options, but some, for expediency, rely on simplification of existing data structures and/or flow mechanisms. We present here the development of a comprehensive multimechanistic (desorption, diffusion, and convection), multiporosity (organic materials, inorganic materials, and fractures), and multipermeability model that uses experimentally determined shale organic and inorganic material properties to predict shale gas reservoir performance. Our multimechanistic model takes into account gas transport caused by both pressure driven convection and concentration driven diffusion. The model accounts for all the important processes occurring in shale systems, including desorption of multicomponent gas from the organics' surface, multimechanistic organic/inorganic material mass transfer, multimechanistic inorganic material/fracture network mass transfer, and production from a hydraulically fractured wellbore. Our results show that a dual porosity, dual permeability (DPDP) model with Knudsen diffusion is generally adequate to model shale gas reservoir production. Adsorption can make significant contributions to original gas in place, but is not important to gas production because of adsorption equilibrium. By comparing triple porosity, dual permeability; DPDP; and single porosity, single permeability formulations under similar conditions, we show that Knudsen diffusion is a key mechanism and should not be ignored under low matrix pressure (Pematrix) cases, whereas molecular diffusion is negligible in shale dry gas production. We also guide the design of fractures by analyzing flow rate limiting steps. This work provides a basis for long term shale gas production analysis and also helps define value adding laboratory measurements.