CO2 Plume Migration During and After Geological Sequestration in Saline Aquifer
| Autor: | Chi-HuiChiu, 邱琪惠 |
|---|---|
| Rok vydání: | 2013 |
| Druh dokumentu: | 學位論文 ; thesis |
| Popis: | 101 Carbon dioxide capture and storage (CCS) is an effective way to reduce carbon dioxide (CO2) emissions into the atmosphere. A deep saline aquifer has the most suitable formation and the largest capability for storing CO2. The purpose of this study is to use analytical and numerical methods to study the front propagation, e.g., plume migration (saturation front) velocity, and migration distance changing with time in two-phase flow in porous media and derived empirical forms of the frontal advance equations. In consider of the complex models in our study, such as 2-dimensional tilted Cartesian and 3-dimensional models, simple analytical solutions were not satifactory. Thus, simulator-generated numerical models were developed to derive empirical equations for estimating plume migration. A variety of operating conditions, such as injection rates, formation permeability and propagation behaviors after stopping CO2 injection were simulated. The numerical simulation method was used in this study to construct 1D linear, 1D radial, 2D Cartesian (horizontal and tilted models are included), and 3D formations (horizontal and tilted models are included). Shock-front saturation, which is obtained from the fractional flow curve, can be used as a criterion for studying CO2 plume distribution at different observation times from numerical results. The major results obtained from this study are：(1) Frontal advance empirical equations for different geometric formations were derived. The propagation results of saturation fronts for linear model shows that there is a linear relationship between frontal advance distance (r) and time (t); for radial and 2D Cartesian models, there exist linear relationships between the square of frontal advance distance (r2) and time (t). (2) Saturation front criteria may affect front propagation velocity. If the chosen front criteria are higher, their propagation velocities are slower. However, if the chosen saturation front criteria are smaller than shock front saturation, their propagation velocity may be the same as that of shock front. (3) In consider of constant injection, the plume-migration velocity is proportional to the injection rate. In variable injection, if total injection rates are the same, then the migration velocity can be estimated by using a single injection rate which is calculated from arithmetical average of variable rates. (4) After the cessation of CO2 injection, the plume propagates farther in the reservoir, especially in updip direction. Plume migrates fast in formation with higher dip angle; in contrast, it migrates slower and even backward in downdip direction because of the buoyancy forces. (5) Both during and after injection, the plumes’s front velocity is unchanged along the ridge-line direction regardless of the dip angle; CO2 migration velocity is proportional to dip angles in updip direction and is inversely proportional in downdip direction. (6) 3-dimensional models consisted of five layers shows that in upper layers, plume propagates outward with time increasing; however, in lower layers, plume migrates backwards due to buoyancy effect. (7) In comparison of wells partially penetrated at the bottom layer and wells fully penetrated, the velocity of the CO2 plume in the top layer is unaffected; but in the central and bottom layers, CO2 plume moves faster in a partially penetrated well than those in a fully penetrated well. (8) In 3-dimensional tilted models, CO2 plume move fast in updip direction, follow by ridgeline direction, and move slow in downdip direction. However, CO2 saturation distribution is complicated in middle layer. Thus, the comparison of migration velocity among three flow directions may vary with different front criteria. |
| Databáze: | Networked Digital Library of Theses & Dissertations |
| Externí odkaz: |
|