Injection/falloff testing of vertical and horizontal wells.

摘要

Our focus is on replacing production/buildup tests by injection/falloff tests in order to estimate reservoir parameters such as permeability and mechanical skin. These tests are economically attractive because they are associated with waterflooding projects. Moreover, they are also used for the purpose of eliminating emissions during well-testing operations. Typically, the reservoir is flooded with water having a temperature considerably below that of the reservoir fluids. This significantly complicates the problem of well test analysis. Not only is the flow in the reservoir governed by the water and oil relative permeabilities, temperature changes induced by injection of cold water also occur in the system. These effects make the associated initial boundary-value problem non-linear. As a consequence, analytical solutions for the pressure are difficult to obtain and the superposition technique usually applied to generate the pressure solution for variable rate problems cannot be theoretically justified.; The intent of this study is to provide a theoretical understanding of the injection/falloff testing of water injection wells. Models for the movement of water injected via a horizontal or a vertical well are developed to generate new approximate analytical solutions for injection and falloff pressures. When incorporating the thermal effects into the analysis, temperature profiles in the reservoir during injection and falloff periods are also constructed analytically. The accuracy of results from our approximate solutions is checked by comparing them to solutions generated from a black-oil reservoir simulator.; The final objective of this work is to provide a practical analysis technique for injection/falloff testing of water injection wells. We demonstrate that our analytical solutions can be used to estimate reservoir and well properties. Injection/falloff data are analyzed using non-linear regression with our approximate analytical model used to construct the predicted pressure response. This approach allows us to estimate important parameters such as the absolute permeability of the reservoir (isotropic and anisotropic), the mechanical skin factor, the length of the horizontal well and the two-phase relative permeability curves assuming a power law model.