Patent Grant
8327936
This application is based on, and claims the benefit of priority to, CA application 2,631,977, filed 22 May 2008, which priority application is hereby incorporated by reference.
This invention relates generally to an in situ thermal process and system for recovering oil from an oil sands reservoir.
Bituminous sands, commonly referred to as oil sands or tar sands, are a mixture of sand or clay, water, and bitumen. While tar sand deposits can be found in a number of different places in the world, the largest tar sand deposits are found in Canada. Most of the Canadian tar sands are located in three major deposits in northern Alberta. Some estimate the Alberta tar sands deposits to contain at least 85% of the world's total reserves of natural bitumen that are concentrated enough to be economically recoverable for conversion to oil at current prices.
Bitumen in its raw state is a heavy viscous crude oil which contains a high amount of sulfur. Because of this high viscosity, bitumen will not flow at reservoir conditions. The two most common bitumen production techniques currently employed are surface mining and in situ thermal recovery.
The largest bitumen deposit in Canada, containing about 80% of Canada's bitumen supply, and the only one suitable for surface mining is the Athabasca Oil Sands along the Athabasca River in Alberta. A smaller deposit is found in the Cold Lake region in Alberta, and is notable for having oil that is fluid enough to be extracted by conventional methods in some places. These Alberta areas are also suitable for bitumen production using known in-situ thermal methods such as cyclic steam stimulation (CSS) and steam assisted gravity drainage (SAGD). These in situ operations involve drilling wells and injecting steam to heat the bitumen allowing it to flow and to be produced from a well.
The use of steam injection to recover heavy oil has been in use in the oil fields of California since the 1950s and is presently being used in several locations in Alberta. CSS, also known as “huff-and-puff” or steam stimulation involves alternately injecting, soaking and producing in a single well. This technique is popular in fields where oil mobility is too low to begin steam flooding immediately. In conventional CSS, steam is first injected into a well at a temperature of 300 to 340 degrees Celsius and at a pressure up to 2000 psi for a period of weeks to months to heat the bitumen (“injection” or the “huff”). The well is then allowed to sit for days to weeks to allow heat to soak into the formation (“soak”). Then, hot water and bitumen are pumped out of the well for a period of weeks or months (“production” or the “puff”). Once the production rate falls off, the well is put through another cycle of injection, soak and production. This process is repeated until the cost of injecting steam becomes uneconomic relative to the money made from producing oil.
SAGD was developed in the 1980s by an Alberta government research center and is now widely used in a number of new in situ projects. In conventional SAGD, a pair of horizontal wells are drilled in the tar sands, with one about 5 meters above the other. Initially, the area around and between the upper well (“injector well”) and the lower well (“producer well”) is warmed up by circulating steam through both wells; during this initial warming up, oil is not produced in commercially significant quantities. Following this, in each well pair, pressurized steam is injected into the injector well, and the heat from the steam melts the bitumen within the heated area or “steam chamber” formed by the steam. The bitumen then flows via gravity into the producer well, where it is pumped to the surface. Each well pair can produce up to 1000 to 1500 barrels per day and are typically spaced 100 to 200 m apart.
SAGD offers several significant advantages over CSS. CSS will typically recover 25-30% of the original bitumen in place over the life of the process, and requires steam to be provided at a higher pressure than SAGD. In contrast, SAGD can recover 60 to 70% of the bitumen in a more efficient manner: for CSS, the steam-oil-ratio (“SOR”) which measures the volume of steam required to extract the bitumen is between 3.0 and 6.0 for CSS and only 2.0 to 3.0 for SAGD. Therefore, significantly less energy, typically in the form of natural gas, is required to generate the steam necessary for the SAGD process.
While SAGD represents a technological advance in certain aspects of recovering oil from tar sands, it is not without its disadvantages. Because this technique relies on gravity for drainage, the process works best in relatively thick and homogeneous clean oil sand reservoirs. CSS in comparison, has been successfully employed in more diverse environments, and is more tolerant than SAGD to variations in reservoir quality.
Another variation of SAGD is known as “Fast-SAGD” has been disclosed, for example, by Polikar et al in H. Shin and M. Polikar, “Review of Reservoir Parameters to Optimize SAGD and Fast-SAGD Operating Conditions”, JCPT Vol. 46, No. 1, January 2007, in U.S. Pat. No. 6,257,334, and by Polikar, M. Cyr, T. J. and Coates, R. M., in “Fast-SAGD: Half the Wells and 30% Less Steam”, Paper No. SPE 65509/PS2000-148, Proc. 4th International Conference on Horizontal Well Technology, Calgary, Alberta (Nov. 6-8, 2000). In Fast-SAGD as disclosed by these publications, an extra single horizontal well is placed between two SAGD well pairs. While the SAGD process is implemented at the SAGD well pairs, steam is injected at higher pressures into the single horizontal well in a cyclic mode in order to help propagate the steam chambers laterally along with the SAGD operation. After several steam cycles at the single well, the single well and SAGD wells are in thermal communication. Then, the single well is converted to production for the remainder of its well life. Meanwhile, steam injection continues into each SAGD injector well to maintain and expand the existing steam chamber, resulting in additional production compared to a convention SAGD operation.
As of the present writing, the proposed Fast-SAGD process has only been simulated and not field tested. The known simulations have only been conducted in idealized reservoirs with uniform properties. Further, the simulations have only been conducted using a two-dimensional vertical cross-section of a reservoir. As a result, the simulations have not taken into consideration the effects of variations in reservoir properties in the cross-section dimensions as well as in the direction of the horizontal wells. It is expected that Fast-SAGD will be problematic in real reservoirs with areal and vertical permeability variations. In nature, every oil sand accumulation will have a certain variation in permeability areally as well as vertically. The CSS wells in the Fast SAGD process operate at pressures that are significantly higher than the SAGD well pairs. The large pressure difference between the CSS and the SAGD wells eventually creates a short circuit between the two wells at which point the process has to be converted to conventional SAGD operation thereby reducing significantly the advantage that would be provided by the CSS well.
With current in situ technologies, the tar sands in Alberta place Canada on par with Saudi Arabia in volume of recoverable oil reserves. Canada is already the largest supplier of crude oil and refined products to the U.S., with over a million barrels per day coming from tar sands. With the recent dramatic increases in oil prices and political volatility in the Middle East, there is strong motivation to develop even more efficient and effective technologies to recover oil from tar sands.
It is an object of the invention to provide a solution to at least some of the deficiencies in the prior art, and in particular to provide an improved in situ thermal process for recovering oil from a tar sands reservoir.
According to one aspect of the invention, there is provided a method of recovering oil from an oil sands reservoir comprising first applying cyclic steam stimulation (CSS) to a series of generally horizontally extending wells in the reservoir; then applying steam assisted gravity drainage (SAGD) to a vertically-spaced well pair in which one well in the well pair is part of the series of wells to which CSS was applied, while producing oil from a single well in the series of wells to which CSS was applied. The single well is laterally spaced from the well pair without any other well therebetween. According to an alternative aspect of the invention, the single well and the well pair can form one group in a repeating group of wells. According to another alternative aspect of the invention, the single well does not form part of a second SAGD well pair, the second SAGD well pair including a well other than the single well and the well pair. According to a further alternative aspect of the invention, in the step of applying SAGD to the well pair, the single well produces oil primarily as a result of SAGD being applied to the well pair.
The method can then further comprise applying a SAGD blowdown to an injector well of each well pair and producing oil from a producer well of each well pair or from the single well or from both the producer well and single well to economic limit.
Both wells of each well pair can be present in the reservoir at the time CSS is applied to the series of wells. In such case, the CSS can be applied to the injector well of each well pair and in which case the producer well is shut in. Alternatively, the CSS can be applied to the producer well of each well pair and in which case the injector well is shut in. The series of wells can comprise an alternating pattern of vertically spaced well pairs and adjacent and offset single wells, wherein the single wells are located at a depth between above the height of the injector well in the well pair by 50% of the vertical spacing between the well pair and below the producer well in the well pair by 50% of the vertical spacing between the well pair.
Alternatively, only one well in each well pair is present in the reservoir at the time CSS is applied to the series of wells. In such case, the other well in each well pair is drilled into the reservoir after CSS is applied but before SAGD is applied. The wells in the series of wells to which CSS was applied can be substantially at the same depth to form an in-line formation; in such case, the wells that are later drilled into the reservoir are each located above a well in the series of wells to form a well pair. Alternatively, the wells in the series of wells to which CSS was applied can alternate between a higher depth and a lower depth to form a staggered formation; in such case, the wells that are later drilled into the reservoir are each located below a well at the higher depth to form a well pair.
The CSS that is applied to the series of wells can comprise multiple cycles, and preferably between one and three cycles. The selected number of cycles will depend on a number of factors such as the properties of the reservoir in which the process is carried out and the well spacing.
In the step of applying CSS to a series of wells, steam can be injected at substantially the same pressure to each well in the series of wells.
a) is a side elevation view of the wells shown in
b) is a side elevation view of the wells shown in
a) is a side elevation view of a series of wells in staggered formation subjected to an initial CSS stage, with future wells to be drilled underneath some of the single wells to form well pairs shown in stippled line.
b) is a side elevation view of a series of wells in an in-line formation subjected to an initial CSS stage, with future wells to be drilled above some of the wells to form well pairs shown in stippled line.
According to one embodiment of the invention and referring to
The CSS-SAGD process can be carried out in tar sand reservoirs that are heterogeneous. For example, the CSS-SAGD process can be used in a reservoir such as the Clearwater formation of the Caribou reservoir at Cold Lake, Alberta. However it is also understood that the CSS-SAGD process can be used in any other tar sand reservoirs having different properties.
The Caribou reservoir is 20 to 32 meters thick and has intermittent layers of shale, breccias and low permeability calcites. The Caribou reservoir has bitumen of 10.9° API gravity with a solution gas-oil ratio of 8.0 m3/m3. The estimated bubble point pressure is 2,650 kPa, the gas specific gravity is 0.65, and the connate water has a total dissolved solid content of 8,889 mg/L.
The wells extend from the surface downwards and then extend generally horizontally into the tar sand reservoir; for the Caribou reservoir, the wells would extend horizontally at about a depth of 425 meters. The horizontal portion of each well extends generally parallel to and are spaced from the horizontally-extending portions of the other wells. The method of drilling such wells are the same as the methods used to drill SAGD and CSS wells, which are well known in the art and thus not described in detail here.
When viewing the wells in cross-section as in
The single wells 14 are located at the same depth as the producer wells 12. Alternatively, the single wells can be located at a different depth, and can be located as high as a depth above the injector wells 10 that is 50% the vertical spacing between the well pairs 10, 12, and as low as a depth below the producer wells 14 that is 50% the vertical spacing between the well pairs 10, 12.
The well configuration of each well pair 10, 12 corresponds to a conventional SAGD well pair. The well configuration of the offset single well 14 corresponds to a conventional CSS well. The well configurations of the wells 10, 12, 14 will depend on the geological properties of the particular tar sand reservoir and the operating parameters of the SAGD and CSS processes, as is well known to one skilled in the art. The spacing between each well pair 10, 12 and offset single well 14 will also depend on the properties of the reservoir and the operating parameters of CSS-SAGD process; in particular, the spacing should be selected such that steam chambers from the injector well of the well pair and the single well can come into contact with each other within a reasonable amount of time so that the accelerated production aspect of the process is taken advantage of. For a medium quality location in the Caribou reservoir, a spacing of 50-75 meters was found to be suitable. The steam chambers come into contact within 3 to 4 years in the Caribou reservoir if the wells are spaced by 75 m.
After the wells 10, 12, 14 are in place, the CSS-SAGD process is carried out to recover oil from the tar sands reservoir. The CSS-SAGD process comprises three key operating stages:
Referring to
According to another embodiment of the invention and referring to
It has been found that in the early years of a well's life cycle, CSS is more productive than SAGD or Fast SAGD in heterogeneous formations. In particular, during this time, CSS has a higher energy efficiency than the Fast SAGD process; the CSS recovery process takes advantage of a variety of recovery mechanisms, including formation re-compaction, solution gas drive, fluid expansion, and a condensate's sensible heat and gravity drainage. CSS is also more tolerant than SAGD of low quality shale and tight streaks.
The initial CSS stage can comprise one or more CSS cycles. Each cycle can take several months to about a year. The number of CSS cycles applied depends on a number of factors including the reservoir properties and well spacing. For instance, in a good quality reservoir, the steam chambers will expand quickly and takes a relatively short period of time to overlap; thus, fewer CSS cycles are required for such wells. For a poor quality reservoir, the steam chambers will expand slowly and takes a relatively long period of time to overlap; thus more CSS cycles are required for such wells. For a good quality tar sands reservoir, one to three CSS cycles should be sufficient for the Initial CSS stage.
A number of operating parameters must be selected during the initial CSS stage. These parameters include steam injection pressure (MPa), maximum steam injection rate (m3/day), steam injection period (days), soak period (days), maximum producer back pressure (MPa), maximum producer steam rate (m3/day), producer total liquid rate (m3/day), and production period (days). These parameters will vary depending on the characteristics of the wells and the reservoir; selection of suitable parameters will be known to one knowledgeable in CSS. For a typical installation at the Caribou reservoir, a suitable maximum injection pressure is 11.5 MPa at 321° C., a suitable steam injection rate is between 600-1000 m3/day, a suitable producer minimum back pressure is 2 MPa, a suitable producer steam rate is 2 m3/day, and a suitable total liquid production rate is 300 to 1000 m3/day.
After CSS was performed for the selected number of cycles, the CSS stage is terminated and the next stage of the CSS-SAGD process begins. In this “SAGD operational stage”, the offset single wells are converted to dedicated production wells (step 60). Then, steam is injected into the injector well 10 of each well pair 10, 12 (step 70); since the surroundings around each well pair 10, 12 were heated during the initial CSS stage, only a relatively short period of time passes before the producer well 12 of each well pair 10, 12 begins producing mobilized bitumen and condensed steam.
The injector wells 10 are injected with steam at a relatively low injection pressure compared to the injection pressures in the Initial CSS stage; for example, in a good quality part of the Caribou reservoir, a suitable injection pressure during the CSS stage is 11.5 MPa and a suitable injection pressure during the SAGD stage is 4 MPa. Other operating parameters that are selected during the SAGD operational stage are similar to conventional SAGD operation and include injector steam injection rate (m3/day, producer operating back pressure (MPa), producer steam rate (m3/day) and total liquid production rate (m3/day) of the production well 12 and the CSS well 14 which operates in a production mode at this stage. The operating parameters will vary depending on the reservoir properties and other factors known to those skilled in the art; for a typical installation at the Caribou reservoir, an injector well steam injection rate is 300 to 1000 m3/day, a suitable production well minimum back pressure is 1 to 2 MPa, a suitable production well steam rate is 2 to 25 m3/day, a suitable total liquid production rate is 300 to 1000 m3/day.
The SAGD stage can be broken down into sub-stages: lateral expansion, overlapping, and downward expansion. During the lateral expansion stage, when steam is continuously injected into the injector wells 10, the steam chamber of the well pairs 10, 12 will expand laterally. A large steam zone is formed above the injector well by the rising steam when the steam chambers of the well pairs and off-set single wells overlap (previously produced during the Initial CSS stage). As a result, dramatic increases in bitumen production and water production rates have been observed, and which are indications of a good combining process.
As can be seen in
The SAGD operational stage takes about nine years in a typical installation at the Caribou reservoir.
This stage takes about two years and is similar to a conventional SAGD blowdown. As the instantaneous steam oil ratio goes beyond the economic limit, all the steam injection will be terminated and the injector wells 10 are shut in (step 80). Hot bitumen will continue to drain to producer wells 12 and single wells 14 and the SAGD chamber gradually cools. The blow down process will be similar to that which follows a conventional SAGD processes.
The advantage of the CSS-SAGD process is that it initiates the production of both the injector well of each well pair and offset single wells with a CSS process conducted at the same pressure, which has been proven to be a successful recovery technique in the Clearwater formation of the Caribou area. The fact that the offset single wells 14 and injector wells 10 in the well pairs 10, 12 inject and produce at the same pressure concurrently, the tendency for the steam to short circuit from one well to the other is eliminated. With the SAGD process starting after a few CSS cycles, the production is accelerated. A mixed SAGD (in well pairs) and steam flooding (in offset single wells) can be considered as a follow-up process to the CSS operation. This provides a greater amount of flexibility and accelerated bitumen recovery in field implementation. Also, and importantly from an economic point of view, it may allow the operator to delay the drilling of the remaining well of SAGD well pairs as shown in
Simulations of the CSS-SAGD process were performed using reservoir simulation models of a well location in the Caribou reservoir in Alberta, Canada. These simulations were compared to simulations of conventional CSS, SAGD, and Fast-SAGD processes performed under the same conditions and using the same simulation model. A summary of the results of these simulations is provided in
The reservoir simulation models were conducted using the Steam, Thermal and Advanced Processes Reservoir Simulator (STARS™) software by the Computer Modeling Group Ltd. (CMG). The reservoir simulation models a location in the Caribou reservoir. The reservoir simulation models are heterogeneous models created in Petrel™ modeling software, and which includes seismic information and well data (well logs and cores) taken from the Caribou reservoir. The grid size selected for all the simulations was 25×2×2 m. A sector model of an average quality reservoir was created which comprised two half single wells in the extremes of the models, one full single well in the middle and two full well pairs in between. The vertical spacing of the well pairs was 5 m. Simulations were carried out with a well spacing at 75 m. The average vertical well depths were 450 m
The models were defined with the following initial rock property distribution parameters; note that permeabilitites and porososities are distributed honoring the real well data and using a geostatistical approach, the figures reported in this table are arithmetic averages.
The reservoir fluid properties of the bitumen are: 10.9° API gravity with a solution gas-oil ratio of 8.0 m3/m3. The gas specific gravity is 0.65 and the water has a total dissolved content of 8,889 mg/L, and the initial pressure was 2800 Kpa at a datum depth of 224 m below sea level.
The “Components” section of CMG's Builder program calculates the PVT parameters of the reservoir. A reservoir temperature of 15° C. and maximum pressure of 15,000 Kpa were used to calculate the PVT parameters. The “Quick Fluid Model” option and the “Blackoil” correlations were used to generate a blackoil type PVT that was then converted into a thermal PVT using the “Import BlackOil” option in Builder (STARS). The solution GOR is 8.0 m3/m3 at initial reservoir temperature and pressure. The oil formation factor at bubble point pressure and reservoir temperature is 1.018 m3/m3.
Another parameter of the simulation is the oil viscosity of the reservoir; in general, oil viscosity reduces with an increase in temperature. The oil viscosity varies from 143557.8 cP at 15° C. (reservoir conditions) to 2.8 at 300° C. (steam injection temperatures). A number of oil samples from stratigraphic wells in the Caribou area were tested and analyzed. For live oil viscosity, the simulator used the dead oil viscosity, a pseudo-solution of gas viscosity for the dissolved gas, and a logarithm mixing rule.
Another parameter of the simulator is the oil/water and oil/steam relative permeabilities of the reservoir. Relative permeability measurements were taken in the laboratory on Caribou Lake cores. Relative permeabilities are affected by temperature changes in a rock system under thermal recovery; the general trend is that the residual oil saturation decreases and the irreducible water saturation increases with an increase in temperature. Relative permeability curves were generated for three different temperatures (15° C., 175° C. and 335° C.)
The dilation option of STARS was utilized with following parameters:
Rock Mechanics Model: Dilation-Recompaction
As there was an expectation that horizontal fractures would be preferentially induced by cyclic steaming at fracture pressures in the reservoir, horizontal permeability multiplier of 50 was applied to the fracture layer in the STARS program.
The simulations were carried out with an objective to confirm that the CSS-SAGD process is feasible and provides the expected advantages over the conventional CSS, SAGD and Fast-SAGD processes. In this connection and referring to
The four processes were carried out until either the instantaneous SOR reached 7 or cumulative SOR reached 4 (“economic limits”).
Results are shown under the “Cumulative” columns in
Referring to the oil production results shown in
It is interesting to note from
One or more currently preferred embodiments have been described by way of example. It will be apparent to persons skilled in the art that a number of variations and modifications can be made without departing from the scope of the invention as defined in the claims.
| Number | Date | Country | Kind |
|---|---|---|---|
| 2631977 | May 2008 | CA | national |
| Number | Name | Date | Kind |
|---|---|---|---|
| 4598770 | Shu et al. | Jul 1986 | A |
| 5273111 | Brannan et al. | Dec 1993 | A |
| 5318124 | Ong et al. | Jun 1994 | A |
| 5417283 | Ejiogu et al. | May 1995 | A |
| 5803171 | McCaffery et al. | Sep 1998 | A |
| 5860475 | Ejiogu et al. | Jan 1999 | A |
| 6257334 | Cyr et al. | Jul 2001 | B1 |
| 6662872 | Gutek et al. | Dec 2003 | B2 |
| 20070295499 | Arthur et al. | Dec 2007 | A1 |
| 20080087425 | Hsu et al. | Apr 2008 | A1 |
| Number | Date | Country |
|---|---|---|
| 1304287 | Jun 1992 | CA |
| 2251157 | Apr 2000 | CA |
| 2277378 | Jan 2001 | CA |
| 2325777 | May 2002 | CA |
| 2330094 | Jul 2002 | CA |
| 2574336 | Aug 2007 | CA |
| 2549614 | Dec 2007 | CA |
| 2593585 | Jan 2008 | CA |
| Number | Date | Country | |
|---|---|---|---|
| 20090288827 A1 | Nov 2009 | US |
