Patent Application
20140083692
The present disclosure relates to methods for improving recovery of hydrocarbons from subterranean formations. More specifically, the disclosure relates to a method of controlling the fluid interface level above a horizontal producer well to effect the inflow of oil-bearing production fluids from the reservoir and to avoid breakthrough of gases into the producer well.
Gravity drainage processes are used for extracting highly viscous oil (“heavy oil”) from subterranean formations or bitumen from oil sand formations. For purposes of this patent specification, the general term “oil” will be used with reference to liquid petroleum substances recovered from subterranean formations, and is to be understood as including conventional crude oil, heavy oil, or bitumen, as the context may allow or require.
For heavy oil or bitumen to drain from a subterranean formation by gravity, its viscosity must first be reduced. The Steam-Assisted Gravity Drainage (SAGD) process uses steam to increase the temperature of the oil and thus reduce its viscosity. Other known gravity drainage processes use solvents or heat from in-situ combustion to reduce oil viscosity.
SAGD uses pairs of horizontal wells arranged such that one of the horizontal wells, called the producer, is located vertically below a second well, called an injector. The vertical distance between the injector and producer wells is typically 5 meters (5 m). The horizontal section of a SAGD well is typically 700 m to 1500 m long. For SAGD projects in the Athabasca oil sands in Alberta, Canada, the depth of the horizontal section is typically between 100 m and 500 m from the surface. Bitumen recovery from the oil sands is accomplished by injecting steam into the injector wellbore. Steam is injected from the injector wellbore into the hydrocarbon-bearing formation, typically through slots or other types of orifices in the injector wellbore liner. The steam permeates the formation within a region of the formation adjacent to the injector well; this steam-permeated region is referred to as a steam chamber. As steam is continuously injected into the formation, it migrates to the edges of the steam chamber and condenses at the interface between the steam chamber and the adjacent region of the bitumen-bearing formation. As the steam condenses, it transfers energy to the bitumen, increasing its temperature and thus decreasing its viscosity, ultimately to the stage where the bitumen becomes flowable, whereupon the mobile bitumen and condensed water flow down the edges of the steam chamber, accumulating as a “liquid inventory” in a lower region of the steam chamber and flowing into the producer wellbore. The fluid mixture of flowable bitumen and water that enters the producer well is then produced to the surface.
A significant challenge encountered by operators of SAGD well pairs is controlling the inflow distribution of oil and water over the horizontal length of the producer well, or the outflow distribution of steam, solvents, or combustion gases from the horizontal injector well. In many cases, inflow distributions or steam outflow distributions are biased towards one part of the well—for example, the region near the heel of the well (i.e., where the horizontal producer well transitions to a vertical well to the surface) or the region near the toe of the well. This results in less favourable well economics due to ineffective use of injection fluid (i.e., steam), poor bitumen recovery rates, and low recovery factors (i.e., when parts of the reservoir are not produced). The inflow/outflow biasing is influenced by the reservoir geology, which is largely outside the control of the well operator.
Another important factor influencing inflow and outflow distributions is the sand face pressure distribution along the length of the injector or producer well resulting from wellbore hydraulics. In this context, “sand face” refers to the point where flow emerges from the sand pack. In oil sands, the sand packs around the liner and flow emerges from the point where the sand is retained by the liner and flows into the gaps of the sand screen. The well operator has some control over this factor by means of the well completion design. For a typical injector well injecting steam into the formation through a slotted liner, wellbore steam pressures are highest near the heel and decrease towards the toe due to fluid friction pressure losses in the axial direction of the wellbore. Where wellbore pressures are higher at the heel, greater outflows of steam, solvent, or other injected gas are present. To equalize or create preferential outflow distributions, Dall'Acqua et al. have proposed (in International Application No. PCT/CA2008/000135) an injector completion with a tubing string run inside a liner, whereby the tubing string has ports located along its length that are sized and positioned to create a uniform or preferential sand face pressure distribution over the length of the injector well. The pressure distribution could be customized to achieve preferential outflow distributions into reservoirs with varying mobility (due to varying formation permeability, for example).
The experience of SAGD well operators in Alberta has shown that the performance of gravity drainage wells is affected by both injector and producer completion designs. In some cases, the producer completion has been shown to have a more significant effect on well performance. A method of controlling inflow distributions over the length of a long horizontal producer well is needed. Producer well design requires consideration of additional complexities that are not factors for injector well design. The fluid interface level relative to the producer needs to be managed carefully to both maximize production rates and to protect the producer well from breakthrough of injection gases. Breakthrough of steam into the producer will damage the well and/or related facilities, and breakthrough of other injection gases (e.g., light hydrocarbons such as propane and butane) reduces the efficiency of their function to mobilize bitumen.
The fluid interface (i.e., the interface between the liquid inventory and the overlying steam chamber) is characterized by a density contrast between the injection fluid (typically steam) and the produced oil and water. For purposes of this patent specification, the fluid interface level will be alternatively referred to as the “liquid level”. It is preferred to let the liquid level sit a short distance above the producer well to act as a seal preventing steam from entering the producer well. If steam is allowed to enter the producer, the steam is not being used for heating bitumen and the process becomes less efficient. Steam entering the producer well can also carry sand particles at high speeds and cause erosion of the steel liners and tubing strings in the wellbore.
To evaluate the economics of an oil recovery project, an estimate of the recovery rate is required. For conventional oil wells, an inflow performance relationship (IPR) is used to predict the oil recovery rate for the reservoir pressure and bottom hole pressure conditions expected. In this sense, conventional oil production is driven by pressure not gravity. Therefore, IPRs as used for conventional oil wells cannot be applied to gravity drainage projects, so a gravity drainage inflow performance relationship (GIPR) is needed to estimate the economics of the process.
“Thermal Recovery of Oil and Bitumen” (R. Butler, 1997, 3rd edition, printed by GravDrain Inc., ISBN 0-9682563-0-9) presents formulas for predicting SAGD recovery rates for a given liquid head, or difference in height between the top of the steam chamber and the producer well. The calculation is based on a two-dimensional cross-section of the well and reservoir. Two other factors will affect SAGD production rates that are not covered in these calculations. Firstly, Butler's calculation assumes that the liquid level contacts the top of the producer well. In actuality, it is typical for liquid levels to sit above the producer wellbore forming a liquid “trap” that the producer wellbore is submersed in. As bitumen and water flow through the liquid trap to the producer well, pressure loss will occur. Many SAGD operators have observed significant pressure losses in this region, with resultant reduction in actual production rates relative to predicted rates. While exact causes for these pressure losses are not fully known, they are sometime attributed to two-phase flow (relative permeability) effects, plugging of slotted liners, fines migration, or other causes.
Another important consideration for predicting SAGD production rates is that wellbore pressures and temperatures vary along the length of a long horizontal well. This will cause liquid levels, and thus the depth of the liquid trap, to also vary along the length of the well, which in turn will affect the total production rate from the well. Near-wellbore reservoir heterogeneities (i.e., permeability variations close to the wellbore) will also contribute to inflow variations along the length of the well.
The present disclosure teaches methods for predicting or characterizing an inflow relationship that relates the vertical position of the liquid level to the position of a producer well. This inflow relationship is applied to producer completion design to select wellbore tubular and flow control equipment that will influence the pressure profile along the length of the producer well, which will affect liquid levels. The inflow relationship considers a number of parameters to arrive at a liquid level prediction; these parameters include injection pressure and temperature, pressures in the producer wellbore, subcool (i.e., cooling of liquid below its saturation temperature) at the heel of the producer, and the vertical temperature gradient (i.e., due to heat loss rate to the underburden, or formation below the production zone). These parameters can be measured directly or indirectly by temperature and pressure sensors placed in the injector and producer wellbores.
The permeability of a heavy oil or oil sands reservoir is non-uniform, or “heterogeneous”. Areas with high permeability will tend to allow steam and oil to flow more easily through them; thus these areas are more likely to be depleted sooner than areas with low permeability. Commonly used producer completion strategies provide little restriction to inflow from high permeability areas, so it is likely that reservoirs will be depleted non-uniformly over the length of the well. This could lead to ineffective placement or distribution of steam during the life of the well, which would reduce the overall efficiency of the process. The ideal case is for the reservoir to be depleted uniformly.
The present disclosure teaches methods facilitating the design or selection of means to limit liquid inflow into the producer well from high permeability areas and to control flow from areas with different permeabilities based on liquid level to match reservoir delivery rate. For example, methods in accordance with the disclosure can be used:
According to one embodiment of methods in accordance with the present disclosure, wellbore flows can be designed to match reservoir delivery. Using this method to determine production rate provides a basis for confirming the completion design and adjusting the design to maintain the production distribution. In this way, growth of the steam chamber can be promoted to be uniform. Alternatively, custom growth patterns can be promoted to accommodate specific geological settings for optimal recovery. Depleting the reservoir uniformly will promote uniform steam chamber growth. This is particularly beneficial for wells with water or gas caps that “rob” steam from the steam chamber rather than allowing the steam to be used as intended (i.e., for heating bitumen at the edge of the steam chamber).
Liquid level is a function of a number of parameters including injector pressure, formation heat loss rate, production rate, permeability, and producer wellbore pressure. Injector pressures are set by the well operator to be higher than the original reservoir pressure to allow for steam to enter the pore spaces within the formation. Injection pressures are limited by the fracture pressure of the formation, which is a function of well depth and overburden geology. Higher injection pressures allow for higher steam chamber temperatures. The pressure acting down on the liquid at the liquid-steam interface is expected and presumed to be close to the injector wellbore pressure.
Formation heat loss rates are governed by the heat conductivity of the underburden geology below the producer well. For a reservoir with bottom water below the producer well, heat losses may be higher and therefore the vertical temperature gradients will be higher.
Producer wellbore pressure and production rates are linked. As production rates are increased, wellbore pressures will decrease. Pressure losses of oil and water will occur as they travel downwards through the liquid trap. Pressure losses are associated with flow through porous media, typically calculated in accordance with Darcy's Law. Additional pressure losses in the liquid trap can occur due to flow convergence from the liquid trap into the openings on the horizontal liner of the producer, from plugging of openings in the horizontal liner, fines migration, relative permeability effects, or other causes.
The rates at which these temperatures and pressures decrease are generally outside the control of the well designer. However, the well designer can control the wellbore pressures through design of the producer well completion. For example, a conventional producer completion may use 88.9 mm tubing landed at the toe of the well. If this tubing diameter is increased to 139.7 mm, then pressure losses through the tubing will be lower. Wells are often controlled to a subcool at the heel of the well, which is typically between 5° C. to 20° C. Subcool at the sand face will be higher as pressure loss through the tubing results in higher pressures at the sand face. For a well with 88.9 mm tubing higher tubing pressure losses will occur, which will result in higher liquid levels. By contrast, a wellbore with 139.7 mm tubing will have less pressure loss and therefore a lower subcool at the sand face.
The preceding example demonstrates the effect of wellbore pressure on sand face subcool and consequently on liquid level. The same principles can be applied to more complicated wellbores with flow control devices mounted on the tubing string or on the liner. The sizing and positioning of flow control devices in the wellbore will affect the direction and magnitude of flow at different points in the wellbore, thus affecting the wellbore pressures.
To maximize production, liquid levels can be designed to be as close to the producer wellbore as possible without causing steam breakthrough. Lower liquid levels will provide greater head pressure in the steam chamber to drive gravity drainage to the sump (liquid inventory).
An iterative method can be applied to predict the liquid level height for an expected pressure and temperature gradient through the liquid zone and a known production rate and injector-producer pressure differential. This calculation can be applied over the well length to determine a liquid level distribution for different completion scenarios. Producer wellbore completions can be optimized to raise liquid levels in areas where production needs to be restricted, and completions can be designed to lower liquid levels in areas where production needs to be increased.
The Gravity IPR (Inflow Performance Relationship) relates the pressure difference between the steam chamber and the production wellbore to the flow rate into the production wellbore. Developing or characterizing the Gravity IPR involves using temperature measurements from the field to define an analysis boundary encompassing the production wellbore and part of the liquid inventory (i.e., sump or steam trap) surrounding the wellbore. The relationship between pressure difference and inflow rate is then determined using numerical or analytical methods. The Gravity IPR has several unique features when compared to a conventional IPR:
Accordingly, in one aspect the present disclosure teaches a method for characterizing an inflow performance relationship relating the vertical position of the liquid level of a liquid inventory in a steam chamber in a petroleum-bearing formation relative to a horizontal producer well disposed within the formation, comprising the steps of:
In one embodiment of the method, the temperature at the fluid interface is assumed to equal the steam chamber temperature, and the temperatures at locations within the analysis boundary are calculated from the vertical temperature gradient and the distance below the fluid interface.
In another embodiment, the pressure at the fluid interface is assumed to equal the steam chamber pressure, and the sum of the pressure head and the elevation head is assumed to be constant along the analysis boundary.
In a further embodiment, the steam chamber pressure is assumed to equal the saturation pressure corresponding to the measured steam chamber temperature.
The analysis boundary may be assumed to be a cylindrical boundary centred on the producer wellbore and touching the lowest part of the fluid interface. However, methods in accordance with the present disclosure are not limited to this assumption, and alternative embodiments of the method may assume a different shape for the analysis boundary.
The methods may include the additional steps of determining the relationship between the pressure drawdown and the inflow rate at a plurality of temperature drawdowns, and then plotting the inflow rate as a function of inflow temperature for a constant pressure drawdown.
In addition to flowing radially from the fluid interface to the producer well, liquid may flow axially (i.e, parallel to the producer well) through the near-wellbore reservoir. For purposes of this patent specification, axial flow through the near-wellbore reservoir will be alternatively referred to as “crossflow”. The steps comprising the characterization of the gravity IPR—namely, temperature measurements, analysis boundary definition, temperature mapping, and numerical or analytical analysis—also enable accurate calculation of the axial hydraulic conductivity of the liquid inventory and, in turn, the axial flow rate.
Accordingly, in another aspect the present disclosure teaches a method for characterizing an axial flow relationship relating the conditions at two axial locations along a horizontal producer well disposed within a petroleum-bearing formation to the axial flow rate through a liquid inventory surrounding the producer well, comprising the steps of:
In one embodiment of the method, the axial hydraulic conductivity of the liquid inventory between the two locations is taken as the average of the axial hydraulic conductivity at the first location and the axial hydraulic conductivity at the second location.
In another embodiment, when conditions other than the liquid level are approximately equal at the two locations, the axial hydraulic conductivity of the liquid inventory at the first location is assumed to equal the axial hydraulic conductivity at the second location and, in turn, the axial hydraulic conductivity between the two locations.
In another embodiment, the effective axial hydraulic gradient between the two locations is taken as the difference between the liquid level at the first location and the liquid level at the second location, divided by the axial distance between the two locations.
In a further embodiment, the gravity IPR is characterized at plurality of axial locations along the producer well, and an axial flow relationship is characterized for each pair of adjacent locations to create a system of axial flow relationships.
Embodiments of the invention will now be described with reference to the accompanying figures, in which numerical references denote like parts, and in which:
The pattern of steam migration within formation 30, and thus the configuration of steam chamber 70, will vary with a variety of factors including formation characteristics and steam injection parameters. However, as represented by the idealized configuration shown in
A producer well 60 is installed at a selected depth below and generally parallel to injector 50, such that it can be expected to lie within the zone of liquid inventory 80 upon formation of steam chamber 70. Producer well 60 has slots or other suitable orifices to allow the bitumen/condensate mix in liquid inventory 80 to enter producer 60 for production to the surface 10. For this purpose, producer well 60 typically has a liner with narrow slots or other orifices that allow liquid flow into producer 60 while substantially preventing sand or other contaminants from entering producer 60 or clogging the slots or orifices in the liner.
Temperature drawdown=steam chamber temperature−inflow temperature.
Liquid level=temperature drawdown/vertical temperature gradient.
When coupled to a wellbore hydraulic model, the gravity IPR enables the performance of a production well to be evaluated by measuring the inflow temperature along the well to determine when the liquid level is reaching critical levels (i.e., when fluid level rise in portions of the well compromises production efficiency, or when fluid level drop in portions of the well compromises well integrity). More specifically, the gravity IPR provides a basis for:
The gravity IPR also provides a basis for determining reservoir delivery distribution over the length of the steam chamber:
Other analytical methods for describing the inflow performance of the SAGD or any other gravity process can be calibrated using methods in accordance with the present disclosure. For example a conventional IPR inflow performance relationship can be calibrated by determining the drainage radius in the basic IPR equation as a function of inflow temperature. This can provide an even simpler basis for evaluating SAGD inflow performance. One example of such an application would be wellbore hydraulics programs used for analyzing and optimizing completions for SAGD production.
The gravity IPR may be characterized at a plurality of axial locations along the producer well and axial flow relationships developed for each pair of adjacent locations to create a system of axial flow relationships, or axial flow “network”. When included in a wellbore hydraulic model coupled with the gravity IPR, an axial flow network enables improved estimation of liquid level variations over time, based not only on an imbalance between the inflow distribution and delivery distribution, but also on the axial redistribution of liquid from locations with a higher liquid level to locations with a lower liquid level.
Practical applications of an axial flow network include:
It will be readily appreciated by those skilled in the art that various modifications of methods in accordance with the present disclosure may be devised without departing from the scope and teaching of the present invention. It is to be especially understood that the subject methods are not intended to be limited to any described or illustrated embodiment, and that the substitution of a variant of a claimed element or feature, without any substantial resultant change in the working of the methods, will not constitute a departure from the scope of the invention.
In this patent document, any form of the word “comprise” is to be understood in its non-limiting sense to mean that any item following such word is included, but items not specifically mentioned are not excluded. A reference to an element by the indefinite article “a” does not exclude the possibility that more than one of the element is present, unless the context clearly requires that there be one and only one such element.
Relational terms such as “parallel”, “horizontal”, and “perpendicular” are not intended to denote or require absolute mathematical or geometric precision. Accordingly, such terms are to be understood in a general rather than precise sense (e.g., “generally parallel” or “substantially parallel”) unless the context clearly requires otherwise.
Wherever used in this document, the terms “typical” and “typically” are to be interpreted in the sense of representative or common usage or practice, and are not to be understood as implying invariability or essentiality.
| Number | Date | Country | |
|---|---|---|---|
| 61492618 | Jun 2011 | US |
| Number | Date | Country | |
|---|---|---|---|
| Parent | PCT/CA2012/000516 | Jun 2012 | US |
| Child | 14093456 | US |
