 |
| Titel |
Modeling and comparative study of fluid velocities in heterogeneous rocks |
| VerfasserIn |
Ferdinand F. Hingerl, Konstantin Romanenko, Ronny Pini, Bruce Balcom, Sally Benson |
| Konferenz |
EGU General Assembly 2013
|
| Medientyp |
Artikel
|
| Sprache |
Englisch
|
| Digitales Dokument |
PDF |
| Erschienen |
In: GRA - Volume 15 (2013) |
| Datensatznummer |
250084141
|
|
|
|
|
|
| Zusammenfassung |
| Detailed knowledge of the distribution of effective porosity and fluid velocities in
heterogeneous rock samples is crucial for understanding and predicting spatially resolved
fluid residence times and kinetic reaction rates of fluid-rock interactions.
The applicability of conventional MRI techniques to sedimentary rocks is limited by internal
magnetic field gradients and short spin relaxation times. The approach developed
at the UNB MRI Centre combines the 13-interval Alternating-Pulsed-Gradient
Stimulated-Echo (APGSTE) scheme and three-dimensional Single Point Ramped Imaging
with T1 Enhancement (SPRITE). These methods were designed to reduce the errors
due to effects of background gradients and fast transverse relaxation. SPRITE is
largely immune to time-evolution effects resulting from background gradients,
paramagnetic impurities and chemical shift. Using these techniques quantitative 3D porosity
maps as well as single-phase fluid velocity fields in sandstone core samples were
measured.
Using a new Magnetic Resonance Imaging technique developed at the MRI Centre at UNB,
we created 3D maps of porosity distributions as well as single-phase fluid velocity
distributions of sandstone rock samples. Then, we evaluated the applicability of the
Kozeny-Carman relationship for modeling measured fluid velocity distributions in sandstones
samples showing meso-scale heterogeneities using two different modeling approaches. The
MRI maps were used as reference points for the modeling approaches.
For the first modeling approach, we applied the Kozeny-Carman relationship to the porosity
distributions and computed respective permeability maps, which in turn provided input for
a CFD simulation – using the Stanford CFD code GPRS – to compute averaged
velocity maps. The latter were then compared to the measured velocity maps. For the
second approach, the measured velocity distributions were used as input for inversely
computing permeabilities using the GPRS CFD code. The computed permeabilities were
then correlated with the ones based on the porosity maps and the Kozeny-Carman
relationship.
The findings of the comparative modeling study are discussed and its potential impact on the
modeling of fluid residence times and kinetic reaction rates of fluid-rock interactions in rocks
containing meso-scale heterogeneities are reviewed. |
|
|
|
|
|
|