All producing hydrocarbon reservoirs are subject to the earth's stress anisotropy and to any regional tectonic movements. In addition, the production of fluids from a reservoir formation reduces its pore pressure such that there is an increase in its effective stresses, and dynamic changes can also occur to the horizontal field stresses as a consequence of partial or total lateral restraint to the rock mass. Similarly, the injection of water into a reservoir during water flooding, or gas injection during pressure maintenance, can alter the rock stresses in a similar dynamic manner. Since such changes lead to hysteresis and permeability modification throughout the life of the reservoir, an understanding of such phenomena are of fundamental importance to reservoir performance and reservoir management.

Conventional laboratory methods for studying permeability under simulated burial conditions involve hydrostatic stresses, or sometimes anisotropic triaxial loading in which two of the principal stresses are equal. However, such hydrostatic and triaxial conditions are very rarely encountered in real reservoirs, and true-triaxial stresses are more representative of actual behaviour encountered in the field. Whilst this limitation may not be of such importance when considering strength and deformation characteristics in reservoir rocks, it can be of significance when considering transport characteristics such as permeability or ultrasonic wave propagation. Although previous investigators have utilised true-triaxial stresses on cylindrical and prismatic samples, such studies have been somewhat restricted by the limited deviatoric stress magnitudes which could be achieved with these test systems. This is a major limitation when one needs to consider the highly deviatoric polyaxial stresses in proximity to a wellbore, or where significant changes to horizontal field stresses occur as pore pressures change. In this ongoing investigation, laboratory equipment is being utilised which can apply the three principal boundary stresses independently using cubic samples, loaded in closed-loop control, and which can achieve deviatoric stresses in excess of 200MPa and pore pressures to 145MPa. Under such conditions, the equipment permits measurements of various petrophysical and geophysical properties such as permeability to single- and multi-phase fluids (i.e. relative permeability), electrical conductivity, capillary pressure, and ultrasonic P- and polarised S-wave velocities, as well as mechanical parameters such as strength and deformability.

This paper reports on the new experimental set-up and work which has been carried out to examine the effects of pore pressure changes under hydrostatic, triaxial and true-triaxial stress conditions. The study has looked at hysteresis and compression effects on permeability under hydrostatic and true-triaxial stress conditions prior to incorporating this data into a reservoir simulator. In particular, the main objectives of the experimental work to date were to investigate how anisotropic stresses and pore pressures might influence permeability and the prediction of production-induced changes to the reservoir (i.e. to simulate the reservoir depletion under different stress-paths), to investigate permeability hysteresis under in situ stress conditions, to integrate rock mechanics with the important petrophysical property of permeability, and to compare data obtained under different stress conditions (i.e. routine laboratory tests at hydrostatic stresses versus tests undertaken at true-triaxial stresses).

Results to date have shown the importance of petrophysical measurements under true-triaxial stress conditions and have highlighted some important implications with regard to the management of stress-sensitive reservoirs.

You can access this article if you purchase or spend a download.