The work presented in this paper describes the evaluation and stepwise optimization process for a Steam-Assisted Gravity Drainage (SAGD) project using a representative sector model from a field with fluid and reservoir characteristics from an eastern Venezuela formation.Due to the complexity and number of variables involved in the process, SAGD presents multiple challenges from the design and analysis phases to its final implementation. The objective of this investigation was to understand the impact of key parameters in the process specific to the selected area and to understand the effects on the recovery factor in these reservoirs, which have previously produced with primary recovery mechanisms.The study touches upon the effect of the component grouping for fluid characterization. A preliminary work consisted of reducing the original 14 components identified in the existing Pressure/Volume/Temperature (PVT) analysis into 2 and 3 pseudocomponents and comparing the stability and results using both fluid characterizations to attain reasonable running times in the simulation process.Once the fluid behavior was successfully recreated and the model was set up, a sensitivity analysis was conducted using thermal simulation. The parameters analyzed were vertical well spacing, injection steam rate, well flowing pressure, and horizontal length of the well pair. The effect on the oil recovery from the angle of dip in the reservoir and the orientation of the well pair with regard to the direction of dip were also briefly analyzed.The conclusion presents a highly improved configuration for the SAGD well-pair array that resulted in trebling the oil recovery attained by the initial well arrangement.

A key component influencing the final recovery factor of a Steam Assisted Gravity Drainage (SAGD) well pattern is the correct well placement. The optimum spacing and position of the well pairs within the net pay is a function of in-situ oil viscosities, reservoir properties and heterogeneities defining the steam injectivity and the effective heat transfer within the reservoir.The well design parameters are typically optimized with thermal reservoir simulators constructed considering that both injector and producer wells are parallel to themselves, to the formation boundaries and with the well-pairs aligned throughout their entire length. However, due to various operational realities including the precision of the guiding tools that navigate the wells, actual SAGD well pairs in the field are seldom drilled perfectly parallel.The study was directed to wells drilled in thick, heavy oil reservoirs where the combination of very high permeabilities of poorly consolidated sands and reservoir temperatures higher than about 30 o C [86 o F] have the ability to flow cold at reasonable oil rates in spite of having in-situ oil viscosities in the thousands of centipoises. This paper expands the previous work of the authors (2010) analyzing the fact that wells from a SAGD pair in the field are not likely to be precisely parallel and quantifies the effects that this lack of exact parallelism brings to the production rates and overall recovery factors comparing them to the recovery variations that result from small to moderate reservoir heterogeneities.