WO2013166587A1

WO2013166587A1 - Steam anti-coning/cresting technology ( sact) remediation process

Info

Publication number: WO2013166587A1
Application number: PCT/CA2013/000453
Authority: WO
Inventors: Richard Kelso Kerr; Peter Yang
Original assignee: Nexen Energy ULC
Current assignee: CNOOC Petroleum North America ULC
Priority date: 2012-05-08
Filing date: 2013-05-08
Publication date: 2013-11-14
Anticipated expiration: 2014-11-08
Also published as: CN104271878B; CN104271878A

Description

TITLE OF THE INVENTION

Steam Anti-Coning/Cresting Technology ( SACT) Remediation Process BACKGROUND OF THE INVENTION

As illustrated in Figure 1A, many oil reservoirs have an active bottom water zone 20 beneath a net-pay zone containing oil. If oil, particularly high viscosity in-situ oil, is pumped from a vertical well completed in the oil zone, water can cone up to the production well and inhibit production. In terms of production, coning will reduce oil cuts and increase water cuts until it is no longer economic to produce the well. In the industry, the well is said to have "watered off. The mobility ratio of the oil determines the rate and extent of water coning. Typically, when the oil is heavier, the worse the water-coning problem is. As illustrated in Figure 2, the problem may also be exhibited in SAGD for bitumen recovery with bottom water reservoirs.

Attempts have been made to prevent coning/cresting when reservoir characteristics are known. However, these attempts have had limited impact. Examples of attempts include the following:

1) The production well is completed higher up in the net pay zone, so the water cone has to be elongated before the well waters off. This is a temporary fix at best, and extra production is often marginal.

2) As illustrated in Figure I B, a horizontal well is drilled so the pressure drop of pumping is spread over the length of the horizontal well. However, water will eventually encroach to the well and produce a water crest zone 10 of high water saturation. Similar to a vertical well, the well will water off.

3) Oil production rates are minimized to delay or prevent coning/cresting

4) As illustrated in Figure 3, downhole oil/water separator 30 (DHOWS) with downhole water disposal is installed. (Piers, . Coping with Water from Oil and Gas Wells, CFER, June 14, 2005). The downhole device can be a cyclone. This device, however, requires a suitable disposal zone 40 for water, and it works best on light oils with a high density difference between water and oil. This is not practical for heavier oils. 5) As illustrated in Figure 4, a reverse coning system 50 is installed (Piers, 2005). Water 60 and oil 70 are produced or pumped separately in this system to control coning. Again for heavier oils, the water pumping rate to control coning is very large and impractical.

There have also been attempts to limit the coning/cresting when reservoir characteristics are unknown or coning/cresting isn't large enough to justify prevention investments. Known remediation attempts have had limited impact. Examples of these attempts include the following:

(1) Blocking agents are used to inhibit water flow in the cone/crest zones. Blocking agents inc lude gels, foams, paraffin wax, sulfur, and cement. Each of these have been tried with limited success (Piers (2005)), (El-Sayed, et al., Horizontal Well Length: Drill Short or Long Wells?, SPE 37084-MS, 1996).

(2) Another reactive process is to shut in the oil well that has coned/crested. Gravity will cause the cone/crest zone to re-saturate with oil. However, when the oil is heavier, the time for re-saturation can be very long and the benefits can be marginal.

(3) A slug of gas is injected into the cone/crest zone. In the early 1990's, a process called anti-water coning technology (AWACT) was developed and tested in medium/heavy oils (AOSTRA, AWACT presentation, March 1999). The AWACT process involves injecting natural gas (or methane) to displace water, followed by a soak period (Luhning et al, The AOSTRA anti-water coning technology process from invention to commercial application, CIM/SPE 90-132, 1990). Lab tests indicated that the preferred gas (CO2 or CH₄) has some solubility in oil or water (Figure 9). The following mechanisms were expected to be activated.

a. On the "huff part of the cycle or when gas is injected, methane displaces mobile water and bypasses the oil in the cone zone.

b. On the "soak" cycle or when the well is shut-in, methane absorbs slowly into the oil to reduce viscosity, lower interfacial tension, and cause some swelling c. On the "puff cycle or when the well is produced, gas forms ganglia/bubbles that get trapped to impede water flow. As illustrated in Figure 5, this creates a change in relative permeability. Oil cuts are improved and oil production is increased.

However, benefits only last a few years, and the process can only be repeated 5 or 6 times. Table 1 below summarizes AWACT field tests for 7 reservoir types (AOSTRA ( 1999)). Oil gravity varied from 13 to 28 API, and in situ viscosity varied from 6 to 1200 cp. AOSTRA suggested the following screens for AWACT - 1) sandstone reservoir; 2) oil-wet or neutral wettability; 3) in situ viscosity between 100 to 1000 cp; 4) under saturated oil; and 5) greater than 10m net pay.

Table 1 :

AWACT Reservoir Characteristics

South Jenner AWACT Treatment Summary

(Based on 34 treatments evaluated)

( ) numbers in brackets arc negative

* ratio is m^J gas per m³ of cumulative oil production prior lo treatment

Reservoir Characteristics of Other AWACT Treated Pools

"Initial Reservoir GOR

As illustrated in Figures 6 and 7, AWACT was not always a success (Lai et al., Factors affecting the application of AWACT at the South Jenner oil field, Southeast Alberta, JCPT, March 1999). As illustrated in Figure 8, a test on a horizontal well was inconclusive (AOSTRA (1999)). 4) Cyclic CO2 stimulation is also a method to recover incremental oil. (Patton et al, Carbon Dioxide Well Stimulation: Part 1 - A parametric study, JPT, August 1982). As illustrated in Figure 10, process efficacy drops off dramatically for heavier oils.

Because of the limitations of the prior art, there is a need for a remediation process that reacts to the cresting/coning in oil wells, preferably heavier oil wells.

SUMMARY OF THE INVENTION

The following terms and acronyms will be used herein:

AOSTRA Alberta Oil Sands Technology Research Authority

AWACT Anti-Water Coning Technology

UMTAR United Nations Institute for Training and Research

JCPT Journal Canadian Petroleum Technology

CIM Canadian Institute of Mining

SPE Society of Petroleum Engineers

JPT Journal Petroleum Technology

SAGD Steam Assisted Gravity Drainage

GOR Gas to Oil Ratio

OC Oil Cut

ro Relative permeability to Oil

Krw Relative permeability to Water

SACT Steam Anti Coning/Cresting Technology

STB Stock Tank Barrels

SRC Saskatchewan Research Council

HZ Horizontal (well)

VT Vertical (well)

OSR Oil to Steam Ratio

SOR Steam to Oil Ratio

DHOWS Down Hole Oil Water Separator

EOR Enhanced Oil Recovery

REC Recovery

OOIP Original Oil in Place

Because of the need for a cresting/coning remediation process, SACT is a process that adds steam to the cone/crest zone and heats oil in the cone/crest zone and at the cone/crest zone edges. In a preferred embodiment, the steam addition is followed by a soak period to allow further heating of oil and to allow gravity to cause a re-saturation of the cone/crest zone. Preferably after the soak period, the oil well may then be returned to production. Preferably, the SACT process is applied to 1 ) heavy oils where native oil viscosity is too high to allow rapid oil re-saturation of the cone/crest zone, preferably where the viscosity is >1000cp, and 2) bitumen (SAGD) wells.

According to a primary aspect of the invention, there is provided a cyclic remediation process to restore oil recovery from a primary well that has watered off from bottom water encroachment (cone or crest) whereby:

(1) The primary well has a produced water cut in excess of 95% (v/v),

(2) The oil is heavy oil, preferably with in-situ viscosity > 1000 cp,

and wherein said process comprises:

(3) Injection of steam in the cone/crest zone preferably by a steam slug with a preferred volume of 0.5 to 5.0 times the cumulative primary oil production, preferably where said steam is measured as water,

(4) After steam injection is complete, the well is shut in for a soak period,

(5) The well is then produced until the water cut exceeds 95%

In a preferred embodiment of the process the well was previously steamed. Preferably the steam is injected using the existing primary oil production well. In an alternative embodiment, the steam is added using a separate well.

In another embodiment of the process, the primary well is a horizontal well and bottom water encroachment forms a water crest zone beneath the primary well.

In another embodiment, in the event that the primary well is not suitable for steam injection, several substantially parallel horizontal wells may be linked with a separate perpendicular horizontal well completed in the steam crest zone of each of the parallel horizontal wells.

Preferably several of the substantially parallel horizontal wells may be linked at or near the midpoint of the horizontal well lengths, in the crest zone.

In another embodiment, the heavy oil is bitumen (API <10; μ >100,000 cp). In another embodiment, there is provided a cyclic remediation process to restore bitumen recovery from a bitumen well that has watered off from bottom water encroachment (cone or crest) whereby:

(1) The primary well has a produced water cut in excess of 70% (v/v),

(2) Injection of steam in the cone/crest zone preferably by a steam slug with a preferred volume of 0.5 to 5.0 times the cumulative primary oil production, preferably where said steam volumes is measured as water volumes,

(3) After steam injection is complete, the well is shut in for a soak period,

(4) The well is then produced until the water cut exceeds 70%.

In another embodiment, the bitumen production well is used for steam remediation injection.

In another embodiment, steam injection rates (measured as water) are 0.5 to 5.0 times fluid production rates when the primary well had watered off.

Preferably the steam quality at the steam injector well head is controlled between 50 and 100%. Preferably the well is shut in for a soak period of 1 to 10 weeks. BRIEF DESCRIPTION OF THE DRAWINGS

Figures 1A and I B respectively depict the water cone lean zone of a vertical production well and the water crest lean zone of a horizontal production well

Figure 2 depicts a SAGD Bitumen Lean Zones (Bottom Water)

Figure 3 depicts the prior art DHOWS concept

Figure 4 depicts the prior art Reverse Coning Control

Figure 5 depicts the AWACT effects on Relative permeability

Figure 6 depicts the Incremental AWACT Reserves in pre and post AWACT oil recovery Figure 7 depicts the Frequency distribution of incremental oil following AWACT

Figure 8 depicts oil production and oil cut history of horizontal wells pre and post AWACT Figure 9 depicts the AWACT laboratory tests and water-oil ratios versus time of various gases Figure 10 depicts the stimulation of CO2 of Oil Wells versus oil viscosity

Figure 1 1 depicts the injection of steam via a steam string for SACT according to an embodiment of the present invention Figure 12 depicts the injection of steam via a separate steam injector for SACT according to an embodiment of the invention

Figure 13 depicts SACT well for Crested Heavy Oil Wells

Figure 14 depicts SAGD partial coning/ cresting

Figure 15 depicts heat conducted around a hot well

Figure 16 depicts SACT simulation in vertical and horizontal wells according to the present invention

Figure 17 depicts SACT simulation in horizontal wells

Figure 18 depicts SACT Scaled Physical Model Steam Injection Rates

Figure 19 depicts SACT Scaled Physical Model Steam Slug Sizes

Figure 20 depicts SACT Scaled Physical Model Water Cut Offs

Figure 21 depicts SACT Scaled Physical Model Horizontal Well Lengths

DETAILED DESCRIPTION OF THE INVENTION

SACT is a remediation process for heavy oil wells (or for SAGD) that have coned or crested due to bottom water encroachment. The process is cyclic and has the following phases:

(1) The primary production well is shut-in due to high (or excessive) water cuts from bottom water encroachment (coning or cresting).

(2) Steam is injected into the cone or crest zone with at least a sufficient volume to displace the bottom water in the cone/crest zone.

(3) The well is shut-in to soak for a period of time (weeks-months). This allows heat from the steam to be conducted to oil in/near the cone/crest zone, reducing the oil viscosity by heating and allowing the oil to re-saturate the cone/crest zone by gravity.

(4) The well is put back on production.

(5) The process can be repeated.

One of the issues for a conventional heavy oil production facility is that primary production wells are not designed for steam injection. The production wells can be damaged by thermal expansion, and the cement isn't designed for high temperature operations. This problem can be mitigated by one of the following options: (1 ) As illustrated in Figure 1 1, the use of an injection string 80 with separate tubing (and insulation) for steam 90 injection to minimize the heating of the primary well 1 10; or

(2) As illustrated in Figure 12, drill and thermally complete a separate steam injection well 100 for remediation of a single well 130; or.

(3) As illustrated in Figure 13, drill and thermally complete a separate steam injection well 100 linked to several wells 140 150 160, allowing for simultaneous remediation.

Referring to Figure 1 1 , an injection steam string 80 with separate tubing and insulation to minimize the heating of the primary well 1 10 is shown. The well in this instance may be vertical or horizontal.

Referring to Figure 12 a separate steam injection well 100 is used to inject steam in to the water cone 120 according to the present invention. In this Figure, a vertical well configuration is shown for use with a single primary production well 130.

Referring to Figure 13 a SACT steam injector horizontal well 100 is linked to a plurality of horizontal producing wells 140, 150 and 160 to ensure crested heavy oil wells are simultaneously remediated according to the present invention.

Bitumen SAGD is a special analogous case for SACT process applications. If the SAGD project has an active bottom water 20, we can expect that the lower SAGD production well will cone/crest eventually (Figure 2). Bitumen (<10API, >100,00cp in situ viscosity) is heavier and more viscous than heavy oil (1000 to 10,000cp), but after bitumen is heated it can act similarly to heavy oil.

If bitumen is above an active bottom water, SAGD can, theoretically, produce bitumen without interference from bottom water, if process pressures are higher than native reservoir pressure, if the pressure drop in the lower SAGD production well doesn't breach this condition, and if the bottom of the reservoir (underneath the SAGD production well) is "sealed" by high viscosity immobile bitumen underneath the production well. But, this is a delicate balance for the following reasons: (1) Steam pressures can't be too high or a channel may form allowing communication with the bottom water. Subsequent fluid losses can, at best, reduce efficiency and at worst, shut the process down. Water production will be less than steam injection.

(2) The initial remedy to this is to reduce pressures. But, steam pressures can't be too low or water will be drawn from the bottom water zone into the production well (coning/cresting). Water production will exceed steam injection. Also, one of the process controls for SAGD is sub-cool (steam trap control) assuming the near- well bore zone is at saturated steam temperature. This control will be lost when bottom water breaches the production well.

(3) As illustrated in Figure 14, if the SAGD reservoir is inhomogeneous or if the heating pattern is inhomogeneous, the channel or the cone/crest can be partial and the problem can be accelerated in time.

(4) Initially, cold bitumen underneath the production well will act as a barrier to prevent channeling, coning or cresting. But, as the SAGD process matures, after a few years, the bottom bitumen will be heated by conduction (Figure 15) and in situ viscosity will be similar to heavy oil, with increased chances of channeling, coning and cresting.

Once the production well has coned/crested, the SACT process can be applied. Unlike heavy oil, the SAGD production well has been thermally completed and it can be used as a SACT steam injector.

Again, the SACT process is cyclic with the following steps:

(1) Shut-in the SAGD producer and convert it to a steam injector.

(2) Maintain target pressures in the SAGD steam chamber closer to but slightly above in situ pressures by using the steam injector well.

(3) Inject a slug of steam into the SAGD production well.

(4) Shut in both SAGD wells for a soak period (weeks-months) to allow bitumen to be heated and to re saturate the cone/crest area.

(5) The process can be repeated.

Example Nexen conducted a simulation study of SACT using the Exotherm model. Exotherm is a three- dimensional, three-phase, fully implicit, multi-component computer model designed to numerically simulate the recovery of hydrocarbons using thermal methods such as steam injection or combustion.

The model has been successfully applied to individual well cyclic thermal stimulation operations, hot water floods, steam floods, SAGD and combustion in heavy hydrocarbon reservoirs (T.B. Tan et al., Application of a thermal simulator with fully coupled discretized wellbore simulation to SAGD, JCPT, Jan. 2002).

We simulated the following reservoir:

Pressure - 6200 kPa

Temperature - 28 degrees Celsius

Porosity - 33%

Initial water Sat. - 30%

In-situ viscosity - 2000 cp

Oil pay - 16 m

Bottom water - 10 m

HZ well spacing - 75 m

HZ well length - 1000 m

We simulated SACT after primary production coned/crested wells. For a vertical well we used steam slug sizes from 50-200 m³. For horizontal wells we used slug sizes an order-of-magnitude larger.

Figure 16 shows simulation results for SACT and a comparison of horizontal and vertical well behavior. Based on the simulation results, the following is observed:

(1) The primary production period for vertical wells is much shorter than for horizontal wells - about a quarter of the time - until the wells are watered off.

(2) The primary productivity of vertical wells is about a factor of 10 less than for horizontal wells. SACT productivities maintained this ratio. (3) The SACT cycle times are larger for horizontal wells. In the period shown in Figure 16 - about 3 yrs. - we have 1 1 SACT cycles for vertical wells compared to only 3 cycles for horizontal wells.

Figure 17 shows a comparison of SACT for horizontal wells, where the steam injection was applied at the heel and at the mid-point of the wells.

Based on the results shown in Figure 17, the following is observed:

(1 ) Primary recovery factor for a horizontal well is about 9% OOIP.

(2) The SACT process, over a period of 2 years after primary production, recovered an extra 5% OOIP for SACT applied at the heel of the horizontal well and an extra 12% OOIP for SACT applied at the mid-point of the horizontal well. This incremental RF is significant when compared to primary production.

(3) The first cycle of SACT applied to the mid-point of the horizontal well produced a production profile better than the primary producer.

In 1995-96 Nexen contracted SRC to conduct a scaled-physical model test of the SACT process based on the following:

14m oil pay column

16m active bottom water column

32% porosity

4D permeability

3600 cp in-situ viscosity

980 kg/m³ oil density (API=12.9)

28°C, 5Mpa reservoir T,P

150m well spacing, 1200m horizontal well length

Tables 2, 3, 4 and Figures 18, 19, 20, 21 present the results of the studies. Based on the results of these studies, the following was observed:

(1) For horizontal wells, steam slug sizes varied from about 36,000 to 54,000 cubic meters (225 K bbl to 340 K bbl) (Table 2). For vertical wells, steam slug size varied from about 500 to 1 100 cubic meters (3100 to 7000 bbls. At least within the range studied, steam slug size is not very sensitive (Figure 1 )). The slug size ratio horizontal/vertical is about 50-70. (Table 3).

(2) Steam injected rate varied from about 300 to 400 m3/d (1900 to 2500 bbl/d) for horizontal wells (Table 2) and at about 9.3m3/d (60 bbl/d) for vertical wells (Table 3). The horizontal/vertical ratio, defined as the ratio of length of contact with oil portion of reservoir, is from about 30 to 43. Steam injection rate is not a sensitive variable (Figure 18).

(3) The SACT process was tested for 4 to 7 cycles for horizontal wells and 3 cycles for vertical wells.

(4) Recovery factors varied from 25 to 36% for horizontal wells and 36 to 43% for vertical wells (OOIP is much higher for horizontal well patterns).

(5) OSR is the key economic indicator. Horizontal wells SACT OSR varied from 0.73 to 0.95 (SOR for 1 .4 to 1.1 ). Vertical well OSR varied from 0.47 to 0.56. In comparison, a good SAGD process has an OSR =.33

(6) Figure 20 shows water cut offs (when production is stopped) are best at higher levels (90% vs. 50%).

(7) Figure 21 shows better performance for longer horizontal wells (300m vs. 150m) but it is not necessarily at optimum lengths.

Based on the studies and simulations discussed herein, it appears that the SACT process of the present invention works best for heavy oil cone/crests, since heating the zone and the oil can improve oil mobility dramatically compared to light oils.

If the heavy oil is produced using horizontal production wells and crests have formed from an active bottom water, a preferred way to link the well crests is a substantially perpendicular horizontal well about mid-way along the crest. (Figure 13) The well is thermally completed for steam injection.

The steam slug should be preferably 0.5 to 5.0 times the cumulative primary oil production, on a water equivalent basis (ie. steam measured as water volumes). The steam injection rate is determined by injection pressures - preferably no more than 10% above native reservoir pressures at the sand face. Enough time is needed for the steam to heat surrounding oil and the oil to re saturate the cone (crest zone) - based on the above, it is preferably between 1 to 10 weeks after the end of the steam cycle.

The process may be repeated when the water cut in produced fluids exceeds about 95% (v/v).

Some of the preferred embodiments of the present invention are provided below.

(1) Heavy oil (> 1000 cp in-situ viscosity)

(2) Well geometry to connect/link to parallel primary horizontal producers in cresting zone.

(3) Preferred linkage near mid-point of horizontal producers.

(4) Steam slug size limits

(5) Soak period limits

(6) Application to SAGD bitumen producer with bottom water

(7) Cyclic remediation process (not continuous)

(8) Applies to both horizontal and vertical wells

(9) Steam injection rate limits

(10) Steam quality limits

Other embodiments of the invention will be apparent to a person of ordinary skill in the art and may be employed by a person of ordinary ski ll in the art without departing from the spirit of the invention.

Table 2:

Scaled Physical Model Test Results Horizontal Wells

(SRC (1997))

Where (1) primary production used in all cases to establish water crests.

Table 3:

SACT Scaled Physical Model Tests Vertical Wells

SRC(1997)

Table 4:

SACT Scaled Physical Model Tests Vertical vs. Horizontal Wells

(SRC ( 1997))

Claims

1. A cyclic remediation process to restore oil recovery from a primary oil production well that has watered off from bottom water encroachment (cone or crest) whereby:

(a) the primary oil production well has a produced water cut in excess of 95% (v/v);

(b) the oil is heavy oil, with in-situ viscosity > 1000 cp; wherein said process comprises:

(c) injecting a steam slug with a volume of 0.5 to 5.0 times the cumulative primary oil production, with steam volumes measured as water volumes;

(d) shutting in the well is for a soak period after the steam injection is complete; and

(e) producing the well until the water cut exceeds 95%.

2. The process according to claim 1, where the primary oil production well has been previously steamed.

3. The process according to claim 1 , where the steam is injected using the existing primary oil production well.

4. The process according to claim 1, where the steam is added using a separate well.

5. The process according to claim 1, where the primary oil production well is a horizontal well and bottom water encroachment forms a water crest zone beneath the primary oil production well.

6. The process according to claim 5, where the primary oil production well is not suitable for steam injection and several substantially parallel horizontal wells are linked with a separate substantially perpendicular horizontal well completed in the steam crest zone of each of the substantially parallel horizontal wells.

7. The process according to claim 6, where the separate substantially perpendicular horizontal well is linked at or near the midpoint of the horizontal well lengths, in the crest zone.

8. The process according to claim 1, where the heavy oil is bitumen (API <10; μ >100,000 cp).

9. A cyclic remediation process to restore bitumen recovery from a bitumen production well that has watered off from bottom water encroachment (cone or crest) whereby:

(a) the bitumen production well has a produced water cut in excess of 70% (v/v);

(b) injecting a steam slug with a volume of 0.5 to 5.0 times the cumulative bitumen, with steam volumes measured as water volumes;

(c) shutting in the well for a soak period after the steam injection is complete; and

(d) producing the well until the water cut exceeds 70%, wherein bitumen is an in-situ hydrocarbon with < 10 API gravity and > 100,000 cp. in-situ viscosity.

10. The process according to claim 9, where the bitumen production well is used for steam remediation injection.

1 1. The process according to claim 9 where steam injection rates (measured as water) are 0.5 to 5.0 times fluid production rates when the primary well had watered off.

12. The process according to claim 9 where steam quality at the steam injector well head is controlled between 50 and 100%.

13. The process according to claim 9 where the well is shut in for a soak period of 1 to 10 weeks.