A screening study and subsequent chemical EOR application pilot strategy for a complex, low-permeability waterflood is presented. Our focus has been on developing appropriate field application options allowing flexibility of operation against a background of reservoir complexity and uncertainty. Australia's Barrow Island Windalia reservoir, the nation's largest onshore waterflood, was developed in the late 1960's. Cumulative oil production to date is approximately 288 MMSTBO. Planning a chemical EOR scheme needs to address the following reservoir and production characteristics:○Highly heterogeneous, very fine grained, bioturbated argillaceous sandstone, high in glauconite;○High porosity (0.28) but low permeability (5 mD with 20 mD+ streaks);○Production and injection necessarily stimulated by induced fractures○Highly saline and hard brine;○Large waterflood pattern volumes (10 MMbbl at 20 acre well spacing). Despite 40 years of production involving water flooding, well-work, and changes in operating philosophy, the nature of the reservoir presents significant uncertainties. These uncertainties flow-on to difficulty in constructing predictive reservoir models. Initial screening recommended that polymers be considered for sweep improvement and conformance control although reservoir complexity presented a challenge. Subsequent laboratory work focused on issues of polymer injectivity, rheology, and retention, in parallel with an assessment of how SCAL properties are measured in the laboratory and related to water flood performance. Dynamic modelling studies have assessed field response and economics for a range of chemical EOR pilot designs. We have focused on developing options for field application of polymers, as opposed to extensive stand-alone laboratory and dynamic modelling studies, in order to address reservoir uncertainties and forecast production response. Results from the proposed polymer pilot flood will allow assessment of further chemical EOR applications and potential field-wide scale up. We propose a mechanism, termed in-depth flow diversion (IFD), which may operate in low permeability, fractured injector water flood. This would allow polymer EOR to operate in lower permeability water flood than currently envisaged. Introduction Australia's Barrow Island Windalia water flood has been on production for 40 years, producing more than 288 million barrels. Windalia is both geologically complex [Hatcher; Williams] and difficult to characterise due to the absence of good seismic reflectors and poor log response. In addition, the Windalia water flood is located in a Class A nature reserve, and hence subject to environmental concerns and responsibilities. Despite these complexities, it is believed there is in excess of 100 million barrels of oil that may be potentially produced through the application of EOR technology.
Producing gas-condensate reservoirs below the fluid's dew-point pressure will lead to an increased condensate saturation in the near wellbore region, which impacts the relative permeability to gas. This phenomenon is known as condensate banking and is a cause of productivity impairment. During a field's operational phase, failure to predict condensate banking behaviour accurately will cause problems with a well's ability to attain production targets. This paper explores the effect of both absolute permeability and condensate gas ratio (CGR) in order to quantify production impairment. To achieve this, three sets of PVT data were characterised. The reservoirs Alpha, Bravo and Charlie have an ideal split (C4-/C5+) CGR of 174.1, 44.7 and 13.9 (stb/MMscf) respectively. For each set of PVT data the absolute permeability has been tested within the range 1 to 1000mD. Both porosity and the relative permeability model were adjusted in proportion to absolute permeability. The range of parameters selected provide coverage of most gas-condensate fields. The PVT data from these reservoirs has been tuned and simulated using a Peng-Robinson twenty-two (22) component dynamic compositional model. Primarily, a single well radial model was used, although implementation of a Cartesian model was also explored for full-field modelling. All models discussed represent an ideal reservoir, consisting of homogeneous properties throughout. Under a steady-state dry gas production rate of 50MMscfpd productivity impairment reached a maximum of 15.4% assuming an absolute permeability of 1mD and an initial CGR of 174.1 stb/MMscf. This low value of 15.4% was assisted by the positive effect of Velocity Dependent Relative Permeability (VDRP) and condensate stripping within the near wellbore region. In contrast, for situations where VDRP does not apply such as low production rates or post shut-down production ramp-up operations, the effects of condensate banking were significant. That is, removing VDRP to simulate these conditions, under the same dry gas production rate of 50MMscfpd a productivity impairment of 93.2% was observed. This result suggests that the most detrimental effect of condensate banking is caused by unsteady-state production operations. It is intended that the findings of this study be applied to wells currently on production to screen for potential condensate banking in later well life.
The method of Dykstra-Parsons, originally developed to model water flood performance, has been adapted to screen reservoirs for their Chemical / Polymer Flood suitability.Firstly, this paper summarises Dykstra-Parsons' Theory alongside an automated method of tuning reservoir layer properties to [pre-chemical] water flood performance.(Methods of determining both layer permeability and oil-phase relative permeability end-points, from water-cut development, are presented.)Secondly, this paper describes the adaptation of Dykstra-Parsons' Water Flood theory to model both chemical flood and post-chemical flood production performance.Permeability adjustment, mobility control, and alkaline-surfactant flood models are discussed. Introduction Dykstra-Parsons' theory, governing the recovery of oil by water flood, has been well documented (ref 1, 2, 3).This theory is versatile but underutilised since the advent and proliferation of finite-difference reservoir simulation.More than fifty years on, however, Dykstra-Parsons remains an efficient yet rigorous method of modelling both water flood and chemical flood performance. Historically, techniques described by authors such as Stiles, Buckley-Leverett, and Welge (ref 2, 3) were integrated in order to model water flood performance via a fractional flow equation.Most importantly these techniques offered a link between: a geological model (albeit one dimensional); an oil in-place volume; and, water flood performance.Certain reservoir conditions must be met for these classical theories to be entirely honoured (ref 2, 3):Steady state flood conditions; voidage-replacement is maintained and the flood is operated above the bubble point pressure (true for many commercial water floods)Diffuse flow - meaning that fluid saturations at any point in the displacement path are uniformly distributed with respect to reservoir thickness. This condition requires either:capillary and gravity forces to be small relative to both high injection and production rates - true for many commercial water floods; or,low rates coupled with the vertical extent of any capillary pressure zone greatly exceeding reservoir thickness (a rarer condition for commercial water floods).The Stiles method of generating pseudo-relative permeability from reservoir layer properties assumes that the mobility ratio is unity.Dykstra-Parsons' theory assumes that reservoir layers flood-out in flow-velocity order. Dykstra-Parsons' theory offers an advantage compared to Stiles in that it is applicable for all mobility ratios (ref 3):Where M < 1 the velocity of frontal advance in each layer will be reduced as the flood progresses, which tends to stabilise the macroscopic flood frontWhere M > 1 the velocity of frontal advance in each layer increases as the flood progresses which promotes instability in the macroscopic flood front. This universal applicability is beneficial when considering that most chemical floods are initiated in order to correct an unfavourable mobility ratio. One misconception is that these classical theories do not apply should a water flood have:significant capillary pressure; or,a vertical permeability and geometry favourable to gravity slumping.This is not the case.In practical terms these theories only require the vertically-acting fluid-redistributive effects, such as capillary pressure and gravity, to be small relative to Darcy or viscous forces in order to remain applicable.It was noted by C. L. Hearn (ref 4) that: "Gravity and capillary forces are neglected and the vertical fluid saturation distribution is assumed to be controlled by viscous flow forces resulting from the vertical permeability variation.The success of stratified reservoir models in simulating waterflood performance indicates that this concept is often reasonable."
In an era of automated workflow-assisted dynamic modelling, Special Core Analysis (SCAL) parameters require updating for each static realisation and evaluation at a quantifiable, probabilistic level-of-certainty. Additionally, SCAL data gaps combined with limited reliable SCAL data drive the need to establish trends and correlations from analogues. SCAL parameters from analogue fields were selected and filtered by depositional environment and laboratory experiment type (centrifuge versus displacement). These analogue SCAL parameters were allocated to statistical bins defined by absolute permeability ranges. Statistical analysis of each SCAL parameter allocated to each permeability bin produced a probability distribution discretised by percentile. Multi-variable linear regression (MVLR) was then implemented to correlate each SCAL parameter, as the response variable, to input variables absolute permeability and percentile. SCAL correlations of reasonable to excellent quality were obtained. The depositional environment was of second order influence in establishing these SCAL correlations. This was due to the selection of core plugs for laboratory analysis from layers of similar quality irrespective of the depositional environment, highlighting the need to select samples characterising a range of lithology and reservoir quality. Centrifuge experiments of water displacing gas were discarded as unreliable due to the compression of the gas phase by the experimental technique. The multi-variable linear regression methodology enabled SCAL parameters to be determined as a function of both absolute permeability and probability. This approach should enable an automated implementation of SCAL parameters within each dynamic model realisation.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.