Transverse permeability measurements of gas shales under replicated in-situ flow conditions: Mathematical modeling and laboratory testing

Abstract

Fast and accurate characterization of unconventional gas reservoirs is of great significance for estimation of gas-in-place and enhancement of gas recovery; however, it remains a big challenge to efficiently and effectively determine the permeability of tight-structure reservoirs in the nano-Darcy scale, especially for those reservoirs with strong gas sorption potential. A novel pressure transient technique is proposed to replicate the in-situ gas flow behavior, and thus to reduce the permeability errors from gas compressive storage, gas sorption, and gas compressibility, which are the three primary error sources widely existing in various permeability measurement techniques. The feasibility of the technique is firstly verified via numerical solutions of a partial differential equation (PDE) model established to closely represent the proposed technique. The results of the numerical study showed that the effects of the error sources on the measured permeability can be minimized and meanwhile, high accuracy and efficiency can be achieved in comparison with the conventional method. A specially designed apparatus for experimental tests via the proposed technique is then used to investigate the stress-dependent permeability of gas shales. The consistency of the experimental results with the findings from the numerical study further validates the applicability of the technique. Considering the complexity of the analytical solution of the PDE model for experimental data interpretation, shale permeability is computed via a new data interpretation approach which is developed by combining the curve-matching and the finite difference method. The experimental results showed that shale permeability is strongly dependent upon the applied stresses and easily affected by gas sorption, gas compressibility and stress inhomogeneity. Without consideration of these factors, the permeability error can reach up to 28.1 % at low stresses. The proposed technique provides a great tool to get insights into the in-situ gas flow behavior and characterize tight-structure gas reservoirs.