Summary
This paper examines the limits of the Walsh and Lake (WL) method for
predicting the displacement performance of solvent flood when miscibility is
not achieved. Despite extensive research on the applications of fractional-flow
theory, the prediction of flow performance under the loss of miscibility has
not been investigated thoroughly.
We introduce the idea of an analogous first-contact miscible (FCM) flood to
study miscibly degraded simultaneous water and gas (SWAG) displacements using
the WL method. Furthermore, numerical simulation is used to validate the WL
solution on one oil/solvent pair. In the simulations, the loss of miscibility
(degradation) is attributed to either flow-associated dispersion or
insufficient pressure to develop the miscibility.
1D SWAG injection simulations suggest that results of the WL method and the
simulations are consistent when dispersion is limited. For the 2D
displacements, the predicted optimal water-alternating-gas (WAG) ratio is
accurate when the permeable medium is fairly homogeneous with a limited
crossflow or is heterogeneous with a large lateral correlation length (the same
size or greater than the interwell spacing).
The results suggest that the accuracy of the WL method improves as crossflow
is reduced. In addition, linear growth of the mixing zone with time is observed
in cases for which the predicted optimal WAG ratio is consistent with the
simulation results. Hence, we conclude that the WL solution is accurate when
the mixing zone grows linearly with time.
© 2012. Society of Petroleum Engineers
View full textPDF
(
5,857 KB
)
History
- Original manuscript received:
10 March 2010
- Meeting paper published:
25 April 2010
- Revised manuscript received:
24 December 2011
- Manuscript approved:
10 January 2012
- Published online:
27 June 2012
- Version of record:
12 September 2012
View Cart
