Relative Permeability Measurements for Post-Waterflood Depressurization of the Miller Field, North Sea
- Peter Naylor (AEA Technology) | Terence Fishlock (AEA Technology) | David Mogford (AEA Technology) | Robert Smith (AEA Technology)
- Document ID
- Society of Petroleum Engineers
- SPE Reservoir Evaluation & Engineering
- Publication Date
- August 2001
- Document Type
- Journal Paper
- 276 - 280
- 2001. Society of Petroleum Engineers
- 1.8 Formation Damage, 5.4.2 Gas Injection Methods, 5.2.1 Phase Behavior and PVT Measurements, 1.6.9 Coring, Fishing, 5.1 Reservoir Characterisation, 5.4.1 Waterflooding, 7.2.1 Risk, Uncertainty and Risk Assessment, 5.2 Reservoir Fluid Dynamics, 5.3.4 Reduction of Residual Oil Saturation, 4.6 Natural Gas, 4.1.5 Processing Equipment, 5.8.8 Gas-condensate reservoirs, 5.5 Reservoir Simulation, 5.6.5 Tracers, 4.1.9 Tanks and storage systems, 4.1.2 Separation and Treating
- 1 in the last 30 days
- 298 since 2007
- Show more detail
- View rights & permissions
|SPE Member Price:||USD 10.00|
|SPE Non-Member Price:||USD 30.00|
This paper describes a determination of critical gas saturations and relative permeabilities relevant to the depressurization of the Miller field. A series of reservoir-condition coreflood experiments and associated numerical simulations is described. Three experiments were conducted with aged Miller core and fluids at 120°C. Each was comprised of a waterflood at about 414 barg, followed by depressurization at different rates. The laboratory data included extensive three-phase in-situ saturation measurements that were used to derive gas relative permeabilities through the simulations.
The rate-dependent critical gas saturations varied between 0.06 and 0.21, and gas relative permeabilities of the order of 0.0001 were deduced. These laboratory results are consistent with published data and suggest that conventional Corey-type gas relative permeabilities are an order of magnitude too large and do not represent the depressurization process.
Significant quantities of hydrocarbons can be left in reservoirs after waterflooding. In those reservoirs with a high solution gas/oil ratio (GOR), depressurization as an improved recovery project can be an attractive possibility.1,2 An assessment of the potential for a depressurization project in the Miller field, North Sea, has been published.3 Risk analysis was used to quantify the economic impact of each uncertainty. Critical gas saturation had the highest impact on the net present value of the project: by varying the critical gas saturation from 0 to 0.15, the peak in gas-production rate was delayed by about 12 years, and the total gas production was significantly reduced.
Many papers have used external gas drive experiments to investigate three-phase relative permeabilities. However, this flow regime is quite different from the in-situ evolution of gas during the depressurization of a volatile oil. Consequently, laboratory studies of depressurization must involve internal gas drive, which is representative of the processes occurring in a reservoir.4,5
Most studies have focused on the critical gas saturation at which gas becomes mobile under virgin conditions. There appear to be fewer laboratory studies of the depressurization of waterflooded oil, and critical gas saturations in the range 0.03 to 0.27 have been observed.6-10 This parameter is believed to be reservoir-specific and dependent upon the depressurization rate.
It is believed that this paper is the first publication of reservoir-condition depressurization experiments involving high-GOR North Sea crude and aged, waterflooded cores. Thus, these may be the most representative measurements of gas relative permeabilities during solution gas drive that have been published to date.
The experiments were conducted in a purpose-built high-pressure rig; the development of laboratory methods is described more fully in Ref. 11. A simplified flow circuit is presented in Fig. 1. It consisted of a vertical core holder, visual cell, separator vessel, and two parallel extraction circuits, all housed within a large temperature-controlled oven. Each extraction circuit comprised a 15-mL vessel into which the visual cell fluids were withdrawn, and a 500-mL vessel into which the nitrogen from the gas spring was collected. Flow rate was controlled by extracting fluid with alternate sides of a twin-barreled, constant-volume extraction pump. The ultra-low depressurization rates required in these experiments were achieved by using the extraction pump in conjunction with a large nitrogen gas spring in the core outlet circuit. This arrangement enabled extraction rates to be as low as 0.0125 mL/hr over a depressurization period of 141 days. Pressures, temperatures, flow rates, produced fluid volumes, and in-situ saturations were measured throughout the experiments.
During coreflooding, it is important that any leak rates are much smaller than the target flow rates. Because the flow rates were relatively low in these depressurization experiments, an extremely high level of leak tightness was required. A great deal of effort was spent to ensure that the leak rates were as low as could reasonably be achieved. The rig was designed with a minimum number of valves and connections in critical parts of the circuit. Extensive leak testing was performed, and it confirmed that leak rates in critical parts of the circuit were as low as 0.0003 mL/hr at full test conditions.
Three-phase in-situ saturation measurements have been conducted throughout these experiments using the gamma-emission technique, which involved labeling the oleic and aqueous phases with gamma-emitting radioactive tracers. This method has been used routinely for nearly 20 years and is described more fully in Refs. 12 and 13. The amount of tracer is too small to affect the phase behavior, but some tracer adsorption was experienced; this is discussed in Ref. 11. The overall uncertainty in in-situ gas saturation was estimated to be approximately ±5% of the saturation measurement.
Core and Fluid Selection.
Three experiments have been conducted for the Miller field using core material supplied by BP plc. All cores were relatively homogeneous, well-consolidated sandstone and two core assemblies were used in this study. Experiment 1 involved a relatively low-permeability core (27 md), while Experiments 2 and 3 used a different core with a higher permeability of 492 md. To obtain an acceptably large pore volume, it was necessary to butt two cores together. The cores for each assembly were selected on the basis of matching porosity, absolute permeability, and pore-size distribution.
There was concern relating to the possibility of calcium carbonate scaling during the depressurization; although this may be an issue in some development plans, it was decided that it might obscure the fundamental mechanisms associated with depressurization. The decision was taken to make synthetic formation brines without the calcium, using a modified live brine recipe.
Neither bottomhole fluids nor separator gas was available for use in these experiments, so the live oil was made from stock-tank oil and synthetic gas. The measured bubblepoint pressure was about 338 barg, and the GOR was between 2,290 and 2,479 scf/STB.
|File Size||150 KB||Number of Pages||5|