Layer | Fill | Outline |
---|
Map layers
Theme | Visible | Selectable | Appearance | Zoom Range (now: 0) |
---|
Fill | Stroke |
---|---|
Collaborating Authors
Summary This paper presents a mathematical model for the scale inhibitor squeeze process. This particular model incorporates the Langmuir adsorption isotherm to describe retention/release of the inhibitor with the reservoir rock. This complements a previous, similar squeeze model based instead on the Freundlich adsorption isotherm. The analytical solution for a simplified version of both models requires only minimal computational effort. Both models calculate the scale inhibitor concentration in the produced brine versus the cumulative produced water volume (and hence the squeeze lifetime). The Langmuir-based model appears to describe inhibitor return data and squeeze lifetime better for reservoirs containing a significant amount of clays and carbonate minerals. The Freundlich-based model better matches field data from cleaner formations such as are found in the North Sea. Similar to the previous model, the Langmuir-based formulation illustrates the importance of the retention/release properties on squeeze lifetime. Parameter studies indicate die total amount of inhibitor injected is the most important design parameter controlling squeeze lifetime. Introduction The scale inhibitor squeeze process is a common technique used to treat production wells that form troublesome water-borne precipitates. Figure 1 illustrates this chemical treatment process. Figure 1 also shows a sketch of a typical response curve. initially, there is no scale inhibitor in the produced water until producing back the overflush volume. Thereafter, there is a spike in chemical concentration representing inhibitor not retained in the formation. Next, the residual concentration gradually decreases as the formation water slowly leaches the retained inhibitor (e.g., via adsorption of precipitation) from the reservoir. Once the produced inhibitor falls to its minimum effective concentration, the production well must be treated again. The cumulative produced water volume to this point is the squeeze lifetime. The objective is to maximize the squeeze lifetime at a minimum cost (minimize amount of inhibitor and pumped volumes). These treatments are largely based on rules of thumb and personal experience. This a particular problem in production wells where squeeze lifetimes are short and the causes for this are unclear. Our objective is to develop a simple mathematical model(s) to describe this process. A good model would permit the operator to predict the behavior of scale inhibitor treatments in order to optimize their cost-effectiveness. Others are developing more sophisticated mathematical models for process. While they incorporate additional important mechanisms for the squeeze process such as multidimensional flow, kinetic effects, and dispersion, they have the disadvantage of requiring complicated numerical solutions. In this paper we will restrict ourselves to simpler models that capture the essence of the process and have easy to apply analytical solutions. A previous mathematical model based upon the Freundlich adsorption isotherm is published. The utility of this model has been demonstrated by fitting and predicting much of the available squeeze data from Ninian Field (North Sea, U.K. Sector). It also has described squeeze data for other North Sea reservoirs. P. 107^
- North America > United States (1.00)
- Europe > Norway > North Sea (0.65)
- Europe > Netherlands > North Sea (0.65)
- (3 more...)
- Water & Waste Management > Water Management > Constituents > Salts/Sulphates/Scales (1.00)
- Energy > Oil & Gas > Upstream (1.00)
- Europe > United Kingdom > North Sea > Northern North Sea > East Shetland Basin > Block 3/8 > Ninian Field > Brent Group Formation (0.99)
- Europe > United Kingdom > North Sea > Northern North Sea > East Shetland Basin > Block 3/3 > Ninian Field > Brent Group Formation (0.99)
- North America > United States > West Virginia (0.44)
- North America > United States > Virginia (0.44)
- North America > United States > Pennsylvania (0.44)
Abstract This study formulates a mass balance model in integral form, for predicting permeability reduction caused by asphaltene deposition in vertical and horizontal wells. The theory accounts for both the effect of radius and time on all the parameters affecting the deposition process. This approach accounts, implicitly, for the effects of pressure, temperature, and fluid composition as a function of position in the reservoir. The model accounts for pore-size distribution, rock original permeability, and asphaltene particle-size distribution. The model is general and can be used to estimate the permeability reduction for both homogeneous and heterogeneous formations. Introduction Optimizing production in fields prone to formation damage caused by asphaltene deposition often requires considerable time and effort. This task depends to some extent on accurately estimating the potential of deposition, and then quantifying the amount of deposition, and its impact on formation damage. Recently, Garrouch and Lababidi have developed and implemented an expert system capable of predicting asphaltene precipitation potential. They formulated and fuzzified rules for asphaltene precipitation caused by solubility incompatibility between crude and solvent, and due to the imbalance between various intermolecular forces between resins and colloidal asphaltene particles. They have collected rules of thumb, pertaining to asphaltene precipitation generated from field experience. Membership functions were then constructed to fuzzify these rules which were grouped into three evaluators. The three evaluators were based on refractive index, solubility, and rules of thumb. Reasoning was carried out using fuzzy logic in which asphaltene precipitation was predicted independently by the three modules, and then an overall module was aggregated. The system was capable of estimating the potential of asphaltene precipitation for a wide range of phase-behavior reservoir conditions, and for a much wider range of live and stock-tank crude compositions than what can normally be handled by a human expert. The advantage of this fuzzy system over deterministic mathematical models was two fold. The system used very basic readily available production data, and did not require an extensive amount of phase behavior data. Moreover, the system presented the prediction results in terms of a confidence level. Numerous models have been proposed based on different microscopic theories for predicting the amount of asphaltene precipitation. They range from steric-colloidal to liquid solubility models based on the Flory-Huggins polymer solution theory. There have been, however, very few attempts to model formation damage caused by asphaltene deposition. Most of the models reported are one-dimentional models based on the plugging and non-plugging parallel pathways approach. Gruesbeck and Collins originally developed their theory to model deposition of fines in porous media. They divided the porous medium into pluggable and nonpluggable pathways. Therefore, they represented the porous medium by two continuous branches such that one is of smaller pores that is susceptible to complete plugging. The nonpluggable pathways, however, cannot be plugged completely. As more deposition acts to reduce the pore-throat diameter, the local speed increases high enough to remove some of the deposit buildup out of the pore space.
- Reservoir Description and Dynamics (1.00)
- Production and Well Operations > Production Chemistry, Metallurgy and Biology > Inhibition and remediation of hydrates, scale, paraffin / wax and asphaltene (1.00)
- Facilities Design, Construction and Operation > Flow Assurance > Precipitates (paraffin, asphaltenes, etc.) (1.00)
- North America > United States > West Virginia (0.44)
- North America > United States > Virginia (0.44)
- North America > United States > Pennsylvania (0.44)
- North America > United States > West Virginia (0.44)
- North America > United States > Virginia (0.44)
- North America > United States > Pennsylvania (0.44)