Multiwell Injectivity for CO2 Storage

Abstract

Abstract A multiwell injectivity model is necessary to assess CO2 injection potential in subsurface formations. This necessity becomes almost ineluctable when CO2 is injected into low-permeability formations, which requires multiple injection wells in order to compensate a minimum injection rate for commercial projects. The pressure interference between the injection wells located in the same reservoir causes significant injection losses. This paper examines well interference when injecting fluids with more than a single injector in a porous medium. Two cases of multiwell injection are studied; water injection and gas injection in a water filled reservoir. In both cases the aim is to examine the degree of well interference caused by the number of wells, formation permeability, well spacing, and total injection rate. A steady-state analytical model obtained using the superposition principle is used for the flow of a slightly compressible fluid of constant viscosity and compressibility in a rectangular homogeneous reservoir of uniform thickness, porosity, and permeability. The wells are represented by fully penetrating vertical line sources with a constant well spacing in a regular pattern. The model is used to calculate the required number of wells for given reservoir parameters and total injection rate. The results are compared with those obtained using single-well and multiwell numerical models. Comparisons show that there is a good match between the multiwell analytical and numerical models for permeabilities higher than 10 mD. It is found that the single-well analytical model underestimates the well numbers compared to the multiwell analytical and numerical models for this range of permeabilities whereas for permeabilities lower than 1 mD it has a good agreement with the numerical model. It is argued that the reason for this should be well interference because the flow in low permeability media is relatively slow, causing less interference between the wells. 1. Introduction Today's growing environmental concern on global warming and climate change has made CO2 sequestration one of the potential solutions to reduce emissions to the atmosphere. CO2 emissions to the atmosphere can be reduced significantly by capturing CO2 at stationary sources and storing it in subsurface geological structures. Research on carbon capture and storage (CCS) is currently a worldwide interest (Benson et al., 2005). A robust model of assessing CO2 injectivity for disposal is essential to predict reservoir behavior and storage costs. Commercial reservoir simulators are valuable tools that help assess the injectivity, but they are expensive and not adequate for a direct use in economics models. Many economics models rely on single-phase, single-well analytical solutions which are mostly obtained for a case where a fully penetrating vertical well is located at the center of a circular reservoir (Ahmed, 2001). Such solutions have been reasonably accurate for the valuation of petroleum reservoirs; however, in the case of CO2 sequestration, these models are not adequate. Research has shown that a relatively high number of wells are required for storage of large volumes of CO2, especially for storage in low-permeability reservoirs (Cinar et al., 2008). When injecting through more than one well the interference between the wells causes injectivity loss and this has to be considered for accurate modelling of storage costs. The injectivity loss depends on a number of parameters such as the number of wells, permeability, differential pressures, well spacing, and injection rate. In this paper, we explore the potential of an analytical multiwell injectivity model to assess the effect of the well interference. The model is simply based on the superposition of pressure drops in space for a quadratic homogeneous reservoir of uniform thickness, porosity, and permeability. The wells are vertical, fully penetrating, and distributed in a square mesh with a constant spacing. The model provides a simple, reasonably accurate and fast tool to evaluate the required number of wells for a given total injection rate of CO2. The results obtained from this model are compared with an analytical single well model and a numerical multiwell model.