Development of a multi-mechanistic, dual-porosity, dual-permeability numerical flow model for coalbed methane reservoirs accounting for coal shrinkage and swelling effects

Most existing coalbed methane (CBM) simulators often treat coal seams as dual-porosity, single-permeability systems, which ignore the effects of water presence in the coal matrix. In this study, a compositional dual-porosity, dual-permeability CBM simulator has been developed. The CBM reservoir is treated as a dual-porosity, dual-permeability system consisting of coal matrix and fracture network. The development of the proposed numerical model incorporates the effects of water presence in the coal matrix and those of coal shrinkage and swelling. The transport of gas follows a multi-mechanistic flow mechanism, which is triggered by pressure and concentration gradients. The proposed simulator was successfully validated against a dual-porosity, single-permeability CBM model (PSU-COALCOMP) and a single-porosity, single-permeability model (GEM). The proposed simulator was also tested against the existing commercial and research CBM simulators for CO2-enhanced CBM recovery process. The simulator was also used in parametric sensitivity studies to investigate the validity of the flow mechanism, the effects of water in the coal matrix and coal shrinkage and swelling conditions. History matching was performed on pure CO2 injection and flue gas injection tests using the actual field data.;The governing flow equations of the proposed simulation model were further used in deriving a material balance equation and a dimensionless form of the transport equation for CBM reservoirs exhibiting homogeneous and isotropic properties. The developed material balance equation was successfully tested and implemented in predicting the reservoir size, production performance and average reservoir pressure using the simulated production data. Lastly, the developed dimensionless production rates and dimensionless time were used in the production type curves application for reservoir characterization. A series of production type curves were generated using the proposed simulator. Several sets of type-curve matching exercises were successfully performed to predict reservoir properties on different systems exhibiting a variety of reservoir and fluid properties.