Canada has huge reserves of oil sands containing heavy oil and bitumen. According to recently announced projects, 2.6 million barrels per day out of 3.5 million barrels per day is planned to be recovered by steam-assisted gravity drainage (SAGD). The engineering design and economic analysis of a SAGD project mainly depends upon its performance prediction. The initial production time after steam injection, oil production rate, steam oil ratio and recovery factor are difficult to calculate because of complexities associated with SAGD process, which are main parameters from the perspective of both engineering and economics for the design of a SAGD project. All of these parameters depend mainly upon relative permeabilities of steam, oil and water in the reservoir. As permeability to any phase of fluid is a function of its saturation, its measurement can be divided into two aspects: (1) the ability to measure the saturation and (2) the ability to determine relative permeability itself. As slight change in enthalpy of steam can make it condense into water, accurate measurement of saturation is almost impossible.

The measured steam-water relative permeability curves resemble those of nitrogen and water. This resemblance is used as basis for 3-phase permeability models for gas, water and oil to be used for steam, water and oil. Stone's model (1970), developed for gas, water and oil, gives freedom to calculate 3-phase relative permeability by using end-point values.

The objective of this study is to improve understanding of multiphase flow in SAGD and better performance prediction of a SAGD process. This study enhances understanding of 3-phase flow in SAGD by introducing a new technique to measure three phase relative permeability. To follow the proposed technique, end point value of steam relative permeability is required. A new experimental technique is also suggested to measure steam relative permeability at residual oil saturation which is a major contribution of this study. The experimental results show that steam relative permeability value at end point, which affects the whole steam relative permeability curve, has significant effects on oil production rate, steam oil ratio and cumulative oil recovery during communication between injection and production wells. It is, also, found that Klinkenberg effect can be ignored in case of steam for reservoir engineering studies. The significance of steam relative permeability is also shown using STARS simulator by showing its significant effect on growth of steam chamber, oil flow rate, steam oil ratio and cumulative oil recovery in case of SAGD. An analysis has been made to show measurement of steam permeability at residual oil saturation as a basis for sound engineering design and economic analysis for the application of SAGD.

The significance of steam relative permeability, at residual oil saturation and consequently the whole curve, between end-point liquid saturations, is also demonstrated for different reservoir and fluid characteristics by using simulation. The results show that steam relative permeability at residual oil saturation, that affects the entire curve, is very important in performance prediction calculations, in any circumstance, related to a SAGD process.