INVESTIGATION OF LIGHT OIL DEPRESSURISATION WITH A DYNAMIC NETWORK MODEL: A STUDY OF RELATIVE PERMEABILITIES

摘要

Hydrocarbon recovery involving reservoir depressurisation is a topic of great interest in the petroleum community at present.However,one of the main problems that reservoir engineers face is the acquisition of reliable data related to relative permeability for input into conventional reservoir simulators.Unfortunately,results from macroscopic core depletion experiments are not easily interpreted,as there is often great uncertainty regarding the physical mechanisms taking place in the core-indeed,the experiments themselves are fraught with difficulties and virtually impossible to undertake at field depletion rates.A clearer understanding of the associated physical phenomena would consequently be highly desirable.In this paper,a pore scale depressurisation network simulator previously used to model heavy oil primary depletion(SCA 2003-11),has been substantially extended in order to focus on the behaviour of very light oils.On the basis of the specific chemical and physical characteristics of such systems(low oil viscosity,high oil formation volume factors and in the present case,very low interfacial tensions),a dynamic network model is presented that takes into account the principal physical processes occurring during pressure depletion of light oils,including gravitational migration,viscous effects and oil shrinkage.In this context,the effect of the high oil shrinkage factors,gravitational and viscous forces on the nucleation,and growth and movement of bubbles is studied for the final prediction of the associated gas-oil relative permeabilities.A number of important findings have emerged from this work.For example,we find that light oil depressurisation is the result of a complex interaction of gravitational forces,capillary effects,oil shrinkage factors and,contrary to expectation,viscous forces.In particular,it is seen that viscous biased bubble growth and viscous mobilization of gas ganglia can predominate at certain times during laboratory depletion experiments.The network simulator is also used to model primary and tertiary experimental depletions.The implications on oil production of the underlying physical processes and the corresponding relative permeabilities are discussed with reference to the current literature.