Abstract
Abstract
The gas-assisted gravity drainage (GAGD) process is being developed to overcome the limitations of, and as an alternative to, the conventional WAG process. In our recent paper (SPE 110132) we have presented the visual model results to demonstrate the feasibility of the GAGD process and the various mechanisms responsible for the high recoveries achieved. In this paper, we present visual and quantitative results from the physical model experiments to demonstrate the various modes of operability of GAGD, its applicability to fractured reservoirs, the effect of oil viscosity and a comparison of its performance with WAG and CGI processes.
A Hele-Shaw type model - consisting of two parallel glass plates (23" x 13" x ¼" in size) with ¼" gap between them filled with Ottawa silica sand - has been used in all experiments with a perforated plastic tube serving as the horizontal production well placed at the bottom of the model. Vertical tubes were placed at different depths in the model to serve as gas injectors. The presence of fractures was simulated by placing cylindrical shaped fine wire mesh tubes into the sandpack. Separate models were built to study the effect of gas injection rate, depth, CGI, WAG, huff-and-puff, toe-to-heel, oil viscosity and wettability.
This paper presents video images of the GAGD process in operation in various modes and discusses the quantitative results of these experiments that led us to conclude that, with the exception of toe-to-heel operation, the GAGD process yielded positive results in all the tests with oil recoveries ranging from 54% to 83% OOIP.
This study demonstrates improved GAGD oil recoveries over CGI and WAG, in fractured model over homogeneous, in oil-wet media over water-wet, thereby signifying the potential for wide applicability of the process in both secondary and tertiary modes.
Introduction
The gas-assisted gravity drainage (GAGD) process, being developed at LSU, attempts to overcome the limitations of the conventional gas injection processes such as the Water-Alternating-Gas (WAG) and Continuous Gas Injection (CGI). While WAG and CGI are based in flooding the reservoir horizontally using vertical wells for gas injection and oil production, the GAGD process attempts to flood the reservoir vertically by injecting gas at the top of the payzone using vertical wells and producing oil from horizontal wells placed near the oil-water contact. The visual physical models built to demonstrate this process and the key mechanisms governing such gravity-stable gas injection for oil recovery have been described in our earlier paper (Mahmoud and Rao, 2007). In this paper we attempt to examine experimentally the range of operability of the GAGD process by constructing visual models to test the performance of GAGD in fractured reservoirs, oil-wet reservoirs, moderately heavy oil reservoirs, in addition to testing the influence of gas injection rate and depth as well as comparing it against the convetional gas injection processes.
EOR in Naturally Fractured Reservoirs
CO2 flooding in naturally fractured rocks has revealed the ability of CO2 to interact with in place fluids between the rock and fractures (Darvish et al., 2006), thereby draining the oil out of the low permeability rock matrix into the high permeability fracture and allowing the CO2 to flood the rock for further drainage. The drainage process is either counter or co-current drainage.
In counter current drainage, the gas is diffused inside in the matrix leading to drainage of the oil into the fractures. On the other hand, co-current drainage is dependent on the viscous displacement of oil in the direction of flow.