Modelling Heavy Oil Viscosity During Thermal Stimulation Using the Free Volume Theory

摘要

Large errors in oil viscosity predictions result in large production rate errors during numerical simulation, due to a first-order dependency of flow rates on viscosity.Questions that arise during thermal simulation studies of heavy oil are(1)which viscosity-temperature model could be sufficiently well based on molecular physics such that viscosity prediction,as a function of temperature,would be possible if only oil composition is available;(2)in which temperature range do viscosity predictions need to be most accurate so as to ensure most accurate recovery predictions–the high or the low endThis work will provide simulation engineers with additional tools and information to address these questions,and more accurately model production of heavy oil under thermal stimulation,by providing a viscosity model based on physical properties of molecular structure and thermodynamic behavior,and additionally,will show that the priority temperature range depends on the thermal recovery method. The Free Volume Model(FVM),based on the concept of the‘available free space to which a molecule can move under shear stress’,is applied in this work to three different heavy oils to calculate viscosity as a function of temperature.The model was first extended for application to extra-long chain molecules up to C64H130 by de la Porte(2012)and de la Porte and Kossack(2014);the present work will provide users with guidelines to estimate the FVT molecular characteristics for a resin-asphaltene heavy pseudo-component. Furthermore,the range of temperatures at which accurate viscosity prediction is essential for more accurate oil recovery predictions is investigated for various recovery methods,such as cyclic steam stimulation(CSS),line-drive steam stimulation(SLD),and steam-assisted gravity-drainage(SAGD). Analytical models are used to predict the sensitivity to the viscosity error in a specific temperature range, followed by numerical simulation to confirm the findings. This research,the first to extend the Free Volume Theory to heavy oil components,will contribute towards more accurate numerical simulation of heavy oil reservoirs.