Transcript
SPE 82224 A Holistic Review of the Water Injection Process
B. Palsson∗, D. R. Davies, A. C. Todd and J.M. Somerville, Heriot-Watt University, Edinburgh, U.K. Copyright 2003, Society of Petroleum Engineers Inc. This paper was prepared for presentation at the SPE European Formation Damage Conference to be held in The Hague, The Netherlands 13-14 May 2003. This paper was selected for presentation by an SPE Program Committee following review of information contained in an abstract submitted by the author(s). Contents of the paper, as presented, have not been reviewed by the Society of Petroleum Engineers and are subject to correction by the author(s). The material, as presented, does not necessarily reflect any position of the Society of Petroleum Engineers, its officers, or members. Papers presented at SPE meetings are subject to publication review by Editorial Committees of the Society of Petroleum Engineers. Electronic reproduction, distribution, or storage of any part of this paper for commercial purposes without the written consent of the Society of Petroleum Engineers is prohibited. Permission to reproduce in print is restricted to an abstract of not more than 300 words; illustrations may not be copied. The abstract must contain conspicuous acknowledgment of where and by whom the paper was presented. Write Librarian, SPE, P.O. Box 833836, Richardson, TX 75083-3836, U.S.A., fax 01-972-952-9435.
Abstract Water injection is an essential part of many modern oilfield development plans. The high costs and often tight economic margins associated with offshore developments require that the chosen waterflood design not only provides an optimum sweep efficiency and reservoir pressure support, to maximise the oil production revenue, but also carries an acceptable level of risk in terms of the project costs and technical uncertainties. The objective of this paper is to show how an optimisation process can be used to maximise the value of the injection water. This paper describes a holistic approach for an economic evaluation of the water injection process, integrating the key technical and economical elements. A six–stage process to evaluate the “Value of Injection Water” is presented. For illustrative purposes, this process is applied to show how a holistic technical–economic evaluation of the water injection operational option for a North Sea field case can be carried out. The study concludes that the injection of a mixture of produced and aquifer water is the preferred source for injection water when applying a matrix injection process into a high quality formation. However, this introduction of the produced water increases significantly the uncertainty in the eventual economic outcome. This process for developing an integrated technical and economic comparison of the options available is a tool by which the operator can decide on a decision on future water injection strategy. It can also assign a quantitative value to each volume of injection water - providing a quantitative framework for making day-to-day operational decisions..
∗
Now with Landsvirkjun, Iceland.
1. Introduction Water injection is a key element in modern oilfield operations. Many offshore oilfield development plans call for water injection into the oil reservoirs for waterflooding (sweeping the oil to the producers) and for pressure support (filling the voidage left by the produced fluids), thus maintaining the reservoir and well bottom–hole flowing pressures above the fluid's bubble point pressure. Other objectives can also become important in specific situations, e.g. control of rock compaction etc. A successful water injection scheme can therefore lead to optimum field development by: maximising overall recovery so that an evenly distributed waterfront sweeps the remaining hydrocarbons towards the producers, accelerating hydrocarbon production by maintaining high reservoir pressure and sweeping oil, rather than water, towards the producers, minimising water production and the associated water handling cost and improving both the environmental and technical profile of the operating company (e.g. by (re)injection of the produced water into the reservoir). An unsuccessful water injection scheme, resulting in limited reservoir pressure support, poor sweep efficiency and excessive water production, can reduce the overall oil recovery and cause high water handling costs. This will have a severe impact on the potential project profitability, as well as adding to the general uncertainty in field performance. Efficient control of the water injection process will have a wider impact on the view government regulators have of the operator’s technical competency, aid in obtaining future exploration licences and avoid the imposition of penalties due to exceeding environmental discharge permits. It will thus have an effect on the operator’s share prices via the company image with investors in the financial market. There are many challenges that have to be considered when a water injection project is planned. They can be strategic, technical and, most importantly of all, economic. It is necessary to acknowledge these different factors and fully understand their impact on the nature of the water injection problem. The aim of this paper is to set the scene for modern water injection management by showing how the value of a water injection operation can be quantified and achieve the
2
SPE 82224
optimum economic performance. The methodology is illustrated with a field case study.
2. Water Injection Economics Normally, a degree of downhole, formation plugging due to contaminants (oil and solid particles) has to be accepted as a practical, economic compromise. Quantification of the degree of damage attributable to a particular type of contaminant in the injection water allows it to be set against the potential loss or delay in oil production due to a reduced injection rate. This can then be balanced against savings in water filtration and deoiling treatment costs. Reliable injectivity prediction methods and an understanding of the uncertainties involved are essential for any water injection planning or operational decisions. E.g. whether an injectivity decline should be accepted (figure 1a), what type and how frequently a stimulation or other remedial method should be used (figure 1b), or if the injection system should be designed differently from start to reduce the risk of injectivity decline (figure 1c). Injectivity
Injectivity
Injectivity
but do not have a better way to determine specific filtration requirements for future developments” [1]. More complex water injection operations, such as in produced water reinjection that have proved to be operationally more problematic than seawater injection and thus present a greater technical risk, should make it, in principle, easier to justify such detailed studies. The modern economic environment requires oil companies not only to consider water injection simply as a “cost” but as an operation generating additional value to the asset. Water injection studies should not simply look at methods to reduce cost, but rather to improve the value of the injection water. A successful waterflooding operation may accelerate hydrocarbon production and possibly increase and extend the plateau rate of the production profile. However, the biggest benefits come through enhanced post-plateau production or secondary recovery, as illustrated in figure 2. Both the earlier production and improved overall recovery will add to the value of the overall field development. The difference in the field value with and without waterflooding represents the value of the injection water.
Q0il (bpd) (a)
Time
(b)
Time
(c)
Time
Figure 1: The water injection scheduling: (a) Should the injectivity decline be accepted and future requirements met with new wells (low cost wells, delay capital expenditure, improve design based on previous experience)? (b) Should the wells be regularly stimulated (backflow, hydraulic fracturing, acid wash or injection of acid, solvents or other chemicals)? What will be the long-term efficiency and the total cost involved? (c) Or should the facilities be designed for better water quality from start? That may increase the up-front expenditure, as well as operating costs, but will result in less risk of operational failure and extensive workovers.
The underlying message is that careful planning of the water injection scheme during the field development stage will most likely be less costly than alterations later in the field life. The economics of these options should be analysed and compared for the particular production circumstances at an early stage in the project cycle when they are all technically viable. They may need to be reviewed at regular intervals as new information on the injection operation become available. In spite of their complicated nature, water injection economics are often treated in the simple way of minimising the injection cost per barrel. In 1984 it was reported that: “Unfortunately it is not straightforward to determine the degree of impairment experienced in heterogeneous layered reservoirs, where water injection is being applied. Such figures can only be determined by carrying out regular surveys of the wells to determine the amount of water being taken by layer. These surveys are time consuming and therefore costly. If water injection does not appear to be a problem then it is difficult to justify such studies. Without them, however, we will continue to carry out and criticise the various laboratory approaches,
Advance production
Increase plateau rate
Maintain and extend plateau
Enhance postplateau production
t (years)
Figure 2: The net present value of an oil field is increased by accelerating, increasing and enhancing hydrocarbon production. This additional value represents the value of the injection water.
The value of the injection water is the key business driver for any waterflood operation and an essential measure of injection efficiency. It is in fact the difference between the incremental revenue and costs, resulting from the water injection operation, calculated at a common point in time. This evaluation of incremental cost and revenue is, however, both complex and the figures carry a large uncertainty, not least due to the interrelationship between these factors. We propose a six step procedure to evaluate the water injection process:
2.1 Key Cost and Revenue Elements The first step is to identify and determine the main economic elements of waterflooding, both the cost and revenue elements. Table 1 provides a very general listing of these elements. The cost elements are clearly more in number, and better defined, as most of them are associated with processes performed at surface over which the operator has some
SPE 82224
3
control. They can be selected with the aim of optimising the overall efficiency. The terms representing the benefits of water injection, such as acceleration of hydrocarbon production and the improved overall field recovery, are both much more “loosely defined” and more difficult to quantify as they are associated with underground uncertainty. Injection Well Costs
Production Well
Cost of injection well; design, drilling, completion and possibly modification of platform
Cost of water production; lifting produced water and handling at surface
Cost of equipment for water treatment and pumping and platform capacity
Cost of produced water disposal
Cost of injection operations; pumping, chemicals, plant maintenance and monitoring Cost of workovers, such as tubing replacement, acid, fracturing etc. Benefits
If PWI, then reduced costs due to surface, or other disposal options
Cost of water related workovers, water shut-off and chemical treatments, e.g. scale prevention Possible “loss” of bypassed oil Accelerated production Improved overall oil recovery
Table 1: A summary of the economical elements of water injection, as related to either injector or producer.
2.2 The Interrelationship Revenue Elements
between Cost
and
The evaluation of the benefits of water injection is a complex process, as discussed previously. An influence diagram is an essential tool to illustrate the interrelationship between the various factors which impact the value of injection water (figure 3). Geologic factors
Improved recovery
Injection conformance Accelerated production
Reservoir pressure support
Platform extension Injection rate/pressure
Value of injection water
Additional capex Water injection facilities
Injectivity
Water source
Added revenue
Wells
Well configuration Waterflood design
Oil & gas prices Improved production profile
Additional opex
The waterflood design objectives are to maximise the value of the injection water through optimising the relationship between the added revenue and the additional costs associated with the water injection, as presented in the influence diagram in figure 3. Potentially, additional savings in water supply and produced water disposal costs will occur in the case of produced water re-injection.
2.3 A Deterministic Cash Flow Model The third step is to construct a cash flow model, including the main cost and revenue elements associated with the waterflood operation (figure 3) and when they occur. Analysis of the cash flow (figures 8-11) reveals different aspects of the project economics, such as the size of the project investment (maximum capital outlay), pay-back period or project profitability. Net Present Value (NPV) is one way of presenting the results of a deterministic cash flow model. By doing that, the option between: (1) building more expensive and operationally complex (produced) water treatment facilities, capable of delivering a “maintenance-free” injectoion well and (2) reducing facility cost while increasing the workover budget associated with an injection well requiring regular workovers. The “Value” lost (due to delayed or lost hydrocarbon production) resulting from the decreased availability of injection water in option (2) – since it will always take a finite time to restore the well to production - has to be included in the comparison. It is always possible to recover the volume of injected water “lost” during the period the well is off-line by installing a larger capacity injection system and well. Extra injection wells could even be drilled – however, these are all expensive and normally less attractive options from an economic point of view. For many operators, this is the final stage for the economic evaluation of the water injection scheme. However, the values of many of the important parameters are not known with any degree of accuracy and the uncertainty has to be addressed, as presented in the following sections.
2.4 Uncertainty Range Sensitivity Studies
Additional costs
Well workovers Plant operation
that can not be affected by the waterflood design are presented with shaded balloons.
Tax regime
Monitoring plant & wells
Figure 3: A simplified influence diagram, drawn during the water injection planning phase to represent the value of injection water. The central role of well injectivity is again illustrated (bold lines). It has a major impact on both waterflood performance (injection rate) as well as the development costs (e.g. number of wells, sizing of water treatment facilities, well maintenance etc.). External factors
and
Deterministic
A deterministic study of the evaluation process addresses the range of possible outcomes for each of the elements involved and quantifies the impact on the final project value. The results can be presented in a spider diagram or in a torpedo diagram, as described in figure 4.
4
SPE 82224
NPV ($mm) Description Production Response
–30
–20
Drilling Capex
Facility Opex
0
10
30
20
40
50
60
70
H
L H
Injection Well OPEX
Producer Well OPEX
–10
L
H
L H
L
H
H
Remedial Opex
Figure 4: An example of Tornado Diagram, presenting the sensitivity of the project value (NPV) to possible variations in the basic water injection cost elements, the elements having the biggest impact at the top. Quantitative examples of the impact of variation of some of these parameters are presented in table 7.
The production response is often the critical business driver for a water injection plan. A good reservoir description is therefore important to allow accurate simulation of the production response to the various injection scenarios.
Evaluation
-
Portfolio theory was originally applied in the stock market to optimise the expected profit from two or more investments, at the same time as minimising the chance of loss through diversification. This methodology is relevant in uncertain oilfield operations, e.g. for selection of field– wide stimulation strategy, well design etc. The potential applications of these methods will be discussed in further details in a future paper.
L L
Base Case
2.5 Probabilistic Injection
Value-of-information methodology is especially powerful for designing a water injection monitoring strategy. It emphasises the fact that information being gathered only has a value associated with it if it has the potential to change a decision on how the Asset is operated [2].
L
H
Platform CAPEX
-
L H
Year of First Injection
be used are: - Decision tree analysis can be used during the planning and operational stages of the project to compare different options. Figure 6 shows how the example given in figure 1 can be formulated.
of
the
Accept decline & drill a new well
Water
Once the range of uncertainties have been identified they can be quantified, or described by a probability distribution curve, e.g. normal, log-normal or a triangular distribution, whose width reflects the uncertainty for this particular variable (figure 5). A Monte Carlo simulation can evaluate the most likely outcome, taking into account the uncertainty in all the important parameters. Various software tools exist for carrying out these calculations.
Injectivity decline
Wait for decline and stimulate
Design a better injection system from start
Figure 6: Decision tree for injection well design.
3. A North Sea Case History
Figure 5: A normally distributed NPV of a water injection project. The cumulative probability curve indicates a 16% chance of negative NPV against 84% change of a positive outcome.
2.6 Advanced Evaluation Methods The final step in the evaluation process is to use more specific and advanced tools to study the individual processes within the overall waterflood design e.g. to compare different operational options to identify the most robust/most profitable process. There is not sufficient room in this paper to describe these methods in any detail, but the key techniques which can
A field case, created from published information and supplemented with information from other similar, representative operating environments (e.g. financial information), is used to illustrate some aspects of the methodology discussed above [3]-[8]. Field A is an oil field located offshore North Sea. The field consists of two main reservoirs, A1 and A2, both of sandstone formations. Total reserves are estimated 53 million cubic meters (315 million bbls) with 27 million believed to be in the A1 reservoir but 26 million in reservoir A2. The plateau rate of 20,000 m3/day (125,800 bpd) was achieved in the first year of production due to extensive pre-drilling. The A1 reservoir is dominated by a relatively homogenous 90100 m thick fluvial-channel sandstone formation with permeability ranging from 1000-4000 md (average around 2000) and good vertical and horizontal communication. The oil zone is on average 50 m thick. The reservoir is split into two areas via a main fault. These two areas exhibit only limited pressure communication. Two high rate water injectors
SPE 82224
5
aim to maintain the reservoir pressure (initially 240 bars) above the bubble point pressure (100 bars), one in each part of the reservoir. The A2 reservoir is the more complex. It consists of extensive, faulted, shallow marine sand layers separated by extensive shale barriers. The reservoir height is approximately 75 m in total with net thickness around 30 m. The individual sand bodies are 1-10 m in thickness with average permeability ranging from 10-200 md, coarsening upwards with thin high permeability streaks (up to 5000 md) at the top. The production wells, located near to the centre of the reservoir, are supported with seven water injectors on the flanks. Table 2 summarises the key information available for Field A. A1 Formation Thick, unconsolidated characteristics fluvial sand
A2 Layered shallow marine sand, separated by shale layers and faulted
Recoverable reserves
27 million m3
26 million m3
Permeability
1000-4000 md
10-200 md (streaks up to 5000 md)
Porosity
21-26%
18-27%
Total thickness
100 m
75 m
Net pay
80 m
30 m
Reservoir pressure
240 bars
215 bars
Bubble point pressure
100 bars
170 bars
Water cut
50-90%
10-50%
Injection well information
2 near-vertical
7 vertical, high angle and horizontal
Average initial flow rate / well
10,000 m3/d
10,000 m3/d
Injection mode
matrix
waterflood induced fractures
Table 2: Summary of the key characteristics of Field A.
Aquifer support is insufficient to maintain reservoir pressure above bubble point and seawater injection was started shortly after production commenced. The water injection capacity is 32,000 m3/d and the facility capacity for the handling of produced water is 18,000 m3/d. The field has now been four years in production and is starting to come off plateau production rate. Various operational and reservoir problems have impacted the waterflood process in both reservoirs. Pump failures, unexpected compartmentalisation and unfavourable transmissibility have all resulted in insufficient water injection volume. The resulting lack of reservoir pressure support has begun to affect the well productivity. In an attempt to achieve the target production rate, the production well inflow pressure has been
reduced below bubble point pressure. This resulted in production with an increased gas–oil ratio and an early breakthrough of formation water, especially in the A1 Reservoir where high water cut production is experienced. Currently, oil production is limited by the surface facility’s water handling capacity. The injection water has not yet broken through in the layered A2 reservoir, but injection conformance is considered a major issue here. The seawater is expected to cause significant scale problems as both reservoirs have formation water rich in barium and strontium. A shallower aquifer water zone has been discovered. This is suitable for injection into both reservoirs since analysis of the aquifer water has shown that scaling problems will not occur if it is mixed with the produced water for re-injection.
3.1 Technical Challenges As stated above, the field has now been four years in production and is starting to come off plateau production. A reservoir management review has been sanctioned to define the strategy for the decline period. The key challenges are: 1.
What source of water should be selected for injection into each reservoir? Continuing the current seawater injection policies and accepting the operational problems in both the injection and production wells is likely to result in insufficient water being injected with the consequent negative impact on oil production (base case). Alternatively, the SWI operation can be upgraded with improved water quality control, monitoring and extensive chemical treatment or drill aquifer water wells or reinject/co-inject produced water?
2.
What injection well completion strategies are available to solve problems resulting from water injection into layered formations?
3.
The water production rate is currently higher than expected and the limited water disposal capacity is restricting production rate. How much is this costing the operators? What is the underlying cause of this problem? Too limited water handling capacity installed? Too high gross fluid production rate? Stronger aquifer than expected or poor injection conformance resulting in high permeable thief zones in the injection well providing a “short circuit” for the injection water to the production well?
All of these challenges are suitable for a techno-economic study, as proposed in this paper. The water injection well completion strategy problem is summarised in table 3 but the remaining part of this paper focuses on selection of injection water source.
6
SPE 82224
Completion Vertical or deviated well, open hole completion
Advantages Low cost option. All layers can be reached and minimum flow restriction through the completion.
Vertical or deviated well, selective perforation or chemical conformance control
Ensures that water enters the tighter zones and sweeps at least close to the injection wells.
Horizontal injector, drilled through the tighter zone only
Maximum control of injection profile.
“Controlled” waterflood (thermally) induced fracturing
Disadvantages The tighter layers are likely to plug up quickly. The water will flow into and cool the higher permeability layers, resulting in thermally induced fracture. Then, the fracture will dominate the injection. Completion will cause flow restriction.
Water source
Advantages
SW
Relatively clean (suspended solids and oil), cold (515°C) – can cause thermal fracturing.
Incompatible with formation water, hence risk of scaling, affecting both injection and production wells.
Moderately cool (27°C), clean water (solids and oil) – can cause thermal fracturing.
High cost of producing AW to surface (water supply wells and lift to surface).
AW
The water is likely to flow through the higher permeability zones through cross-flow, deeper in the reservoir. Expensive and complex well option, both for construction and operation.
AW + PW
Fracture conformance to the Reservoir zone is essential, demanding on an extensive monitoring program and study of rock mechanical properties
Table 3: The advantages and disadvantages of different completion options for injectors in a layered formation.
3.2 Selection of Injection Water Source for the A1 Reservoir When reviewing the injection process, the first aspect to consider is the source of the injection water. Three options are considered relevant here, seawater (SW) which is currently used, aquifer water (AW) and produced water (PW). The advantages and disadvantages of these options are listed in table 4. The injection facilities for the two reservoirs are separated, allowing different sources of injection water to be chosen for the two reservoirs. A reservoir engineering study has recommended a matrix injection process for the A1 reservoir to minimise further water breakthrough. Therefore, the source of injection water for the two water injection wells in the A1 reservoir is of greater concern and is studied in detail here. McCune’s injectivity model was applied to predict injectivity behaviour of the different water sources applied [10]. The results are presented in table 5 and figure 7.
PW disposal required with expensive treatment and/or dedicated disposal wells.
PW disposal still required.
Compatible with formation water.
Contacts only limited number of layers. Selected layers fractured before injection starts into the higher permeability layers, to ensure good injectivity and better injection profile [16], [14].
Disadvantages
Moderately warm (30-60°C depending on AW/PW ratio), viscosity is low and thermal fracturing may be possible. Minimises cost of PW treatment and volume of water for overboard disposal.
High solids loading is likely to cause more formation plugging and more facilities maintenance (e.g. pump wear etc.) than SW and AW. Requires higher fracture pressure than AW and SW due to higher temperature. The ratio of PW will increase later in field life as more and more water breaks through.
Table 4: The advantages and disadvantages of the various available injection water sources. Water Surface Viscosity Injectivity Solids Particle Half-life source temp. Index conc. size SW
10°C
1.17 cp 230 bpd/psi 3 mg/l
1 µm
AW
30°C
0.81 cp 332 bpd/psi 1 mg/l
1 µm 172 days
PW
80°C
0.38 cp 708 bpd/psi 9 mg/l
5 µm
AW+PW 55°C
0.54 cp 498 bpd/psi 5 mg/l
57 days
19 days
4.5 µm 34 days
Table 5: Water quality properties of the different injection water options N.B. The Injectivity Index at time zero (II0) will be lower for the more viscous, lower temperature water sources
As illustrated in figure 7, all the water sources are expected to cause injectivity decline and regular acid stimulation treatments will be required to achieve the target injection volume. The pure aquifer water is expected to have the least impact, requiring only one acid stimulation per year. The higher solids and oil loading of the produced water is expected to require at least four acid stimulation treatments each year to reach the same injected water volume target.
SPE 82224
7
II/IIo 1.0
Cost elements: SW AW AW + PW PW
$2,500,000 $1,750,00 0
Increased well cost (AW producer(s) required)
$0
$6,000,000 $3,000,00 0
Increased treatment equipment cost (AW treatment equipment is required)
$0
$500,000
$250,000
Filter-cake: Permeability: 200 md Porosity: 20% SW: 3 ppm, 1 micron AW: 1 ppm, 1 micron AW/PW: 5 ppm, 4 micron
Increased injection pump cost (need increased pumping power for PW)
$0
$0
$500,000
Incremental Opex/bbl over base case
$0.245
$0.172
$0.125
Incremental "average" pump cost /bbl
$0.004
$0.004
$0.011
Incremental chemical injection (SW: $0.200 scale and corrosion inhibitors and AW: corrosion inhibitors)/bbl
$0.100
$0.075
Plant maintenance/bbl, (SW: removal of scale, AW: corrosion, PW: erosion)
$0.040
$0.020
$0.015
AW lifting ($0.06/bbl AW)
$0.000
$0.047
$0.024
Increased monitoring (water quality, $0.001 injection profile and performance):
$0.001
$0.001
0.7 Permeability: 2 darcy Porosity: 25% Thickness: 80 m Initial rate: 10.000 m3/d
SW
0.5 0.4
AW/PW
0.3 0.2 0.1
PW
0.0 0
100
200
300
400
500
600
700 Days
Figure 7: Injectivity prediction, applying McCune’s method, for the possible injection water sources. The figure also indicates the acid stimulation treatments required to achieve the target injection volume.
The incremental cost elements associated with either continuing with an improved SWI operation (monitoring, water treatment and regular workovers) or implementing either AWI or AW/PWRI are listed in table 6. It is assumed that PWRI requires little additional capital expenditure as treatment facilities for offshore disposal are already in place. However, a significant additional cost for AW supply will be incurred. 3.2.1
AW + PW
$0
AW
0.6
AW
Incremental Capex:
0.9 0.8
SW
Economic Analysis
This case history has been simplified by assuming that all three alternative injection schemes will be equally effective in increasing the average water injection rate from the existing 18,000 m3/d to 20,000 m3/d. The increased water injection rate, associated with higher quality injection water sources, improved monitoring and control of the injection profile, will ensure improved reservoir pressure support and sweep. This will lead to a decrease in the water cut and an increase in the oil production. The incremental oil production from the A1 Reservoir due to the improved pressure support is estimated to be 1,000 m3/d during the first two years, declining 20% per year after that. The income generated by the extra oil sales from the three alternative water injection schemes is assumed to be the same. This is not unreasonable as the facilities are water handling capacity limited and another production well can be produced while an active oil producer is taken off-line for three days every six months to allow an inhibitor squeeze treatment to be carried out. The incremental costs and incremental revenue are presented in figure 8, where a net oil price of US$20 per bbl has been assumed.
Tubing replacements (SW: 4 years, AW: 5 years, PW: 3 years): $200,000 each Acid stimulation per well (SW: 2/year; AW 1 /year; AW+PW: 4/year): $50,000 each SWI: 2 injectors; AWI: 2 injectors and 2 AW producers; AW+PW: 2 injectors and 1 AW producer SW
AW
AW+PW 3
Total volume of produced water processed:
10,000 m3/d
10,000 m /d
10,000 m3/d
Total volume of water disposed overboard:
10,000 m3/d
10,000 m3/d
0 m3/d
Total volume of injection water prepared:
20,000 m3/d
20,000 m3/d
10,000 m3/d
Table 6: Cost elements involved when changing water injection scheme, presented in terms of cost in US$ per bbl of total daily injection rate (20,000 m3/d).
8
SPE 82224
50 SW Incremental Cost AW Incremental Cost AW+PW Incremental Cost Incremental Revenue
30 20 10
3.2.2 0 -10 -20 2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
Figure 8: Cost and revenue for the three injection water sources over the 10 year project lifetime. The incremental revenue decreases with time because of production decline, whereas the injection system operational cost remains constant. After 7 years, the cost of water injection exceeds the incremental revenue due to the injection for the SW and AW injection cases.
Figure 9 presents the cumulative cash flow model for the three injection water sources, based on the incremental cost and revenue assumptions presented in table 6 and figure 8 and discounted at the rate of 10% / year.
Cumulative Discounted Cash Flow [Million $] .
140 120
Value Creation by Selection of Optimum Injection Water Source
In theory, this incremental value of the injection water can be considered as a complex partial differential equation as a function of time and other factors having an impact on the waterflood performance:
∆E w (t ) =
∆NPV (t , Qo (t , PRe s ...)..) ∆W (t , Q w (t , BHIP, PRe s ...)...)
This equation is the core of any integrated study of the value of injection water. It gives the value generated by injecting one extra barrel of water per unit time and therefore, a study of the derivative will give the parameters (such as injection rate, stimulation frequency etc.) leading to optimum water injection operation. The data required for this model will have to be derived from a full-field reservoir simulator study coupled to a field cost model. This would follow the same lines, but be a significant extension of the simpler study reported here.
100 Incremental NPV of AW+PW: $132.6 million
80
9
Value of the injected SW Value of the injected AW Value of the injected AW+PW
8 60 SW Cashflow AW Cashflow
40
AW+PW Cashflow 20
0 2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
Figure 9: The incremental Cumulative Discounted Cash Flow and the incremental Net Present Value of three alternatives for injection water schemes for Reservoir A1. The commingled AW+PW creates the greatest value.
In this example, the commingled injection of AW and PW turns out to be the most profitable option with an incremental Net Present Value of US$132.6 million. It should also be noted that, the “Environmental Good” of avoiding discharge of the produced water into the environment has not been assigned a monetary value. The inclusion of even a nominal “Environmental Value” would have increased the incentive to implement the mixed (AW + PW) injection scheme. The value of the additional water injected each year (Ew,i) can be determined by dividing the annual incremental Net Present Value (∆NPVi) for that year by the annual incremental volume of water injected (∆Wi). Figure 10 presents the incremental value of the three water injection scheme options studied. As can be seen, injection of an extra barrel of produced water and
Value of the Injection Water [$/bbl]
Cost / Revenue [Million $]
40
aquifer water will generate the biggest incremental income and the value of the incremental injection water (2,000 m3/d) exceeds US$8/bbl in year two. Later in the field life, the increased daily injection water volume will result in less and less incremental oil production so that the value created per unit of injection water will decline.
7 6 5 4 3 2 1 0 2001 -1
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
Figure 10: The value of the incremental injection water (2,000 m3/d), equivalent to the incremental Net Present Annual Cash Flow divided by the incremental volume of water injected annually, throughout the ten year project life.
Figure 10 included the value created in the three injection scenarios when the optimum volume of water was injected. The extra oil produced, assumed to be constant, has a large impact on the figures quoted. Figure 11 compares the value created between these three optimum injection scenarios. The improved seawater injection option is now used as the base case against which the efficiency of the aquifer water and mixed aquifer/produced water options can be judged. This figure illustrates the fact that the injection of aquifer/produced water mixture generates US$21 million higher incremental NPV than the improved SWI operation.
SPE 82224
9
25 NPV(AW+PW) - NPV(SW)
NPV [$ million]
20
$21.3 million
NPV(AW) - NPV(SW)
15 $9.4 million
10
“Range” column in table 7 shows that the uncertainty in the various elements is greatest for the produced water case. However, the economic analysis has shown that the benefits of reducing the overall water handling at the platform (produced water and fresh injection water) outweigh the costs and uncertainty in injectivity caused by the more complex produced water properties.
∆
5
0 2001
Impact: Element 2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
-5
Figure 11: The cumulative incremental cash flow generated by injecting either AW or a mixture of AW and PW instead of the improved SWI operation.
3.2.3
Project Risk and Uncertainty
The introduction of PW will however, bring a bigger technical uncertainty into the water injection operation. The high loading of solid particles in the produced water is likely to cause erosion of pumps and other equipment in the water injection system. Production upsets in the separation process could possibly result in water of very low quality (oil, solids) being injected for some period of time, exacerbating the formation damage problem. Production upsets are frequently associated with back production of spent acid following an acid stimulation treatment. Operational procedures will have to be developed that can prevent this type of operational difficulty if the (AW + PW) option is chosen. However, the risk that this could occur was already included in the assumption that four acid stimulation treatments would be required each year to maintain the required injection well capacity. The AWI also carries potential uncertainty, mostly associated with the reliability of the water supply system. This will be exposed to the potentially corrosive untreated AW. However, in many cases field personnel have indicated that their greatest concern is associated with the injection well injectivity. Although the initial, actual SWI performance was somewhat better than predicted with the McCune’s model, it is expected that uncertainties associated with the downhole water quality issues such as scaling and corrosion products might lead to greater injection well formation damage later in the field life. Table 7 presents the impact of the three, most critical uncertainty factors (the top three uncertainty factors on the Tornado plot, figure 4). The uncertainty ranges differ for different injection water sources, reflecting the characteristics of the three options. Ideally, the uncertainty ranges would be based on statistical studies of the waterflood performance in representative, equivalent operational environments. A simpler, deterministic “engineering judgement” based approach has to be applied to generate the distribution of the cost and performance figures when only limited operational experience is available (see next section and figure 12). The
Base case Production response:
SW
AW
AW + PW
Range NPV* Range NPV* Range NPV* P50
100%
111.6
100%
120.8
100%
132.6
High (P10)
200%
273.9
200%
283.1
200%
294.9
50%
30.5
50%
39.6
50%
51.5
150%
86.4
125%
111.4
200%
106.9
75%
124.3
75%
129.6
75%
139.1
-
-
200%
115.4
200%
129.9
-
-
-
-
-
-
Low (P90) Fixed Opex (pumping, chemicals) Wells
High (P10) Low (P90) High (P10) Low (P90)
Table 7: Sensitivity of the economic evaluation to various parameters. This table determines the deterministic uncertainty distribution for the three “critical uncertainty factors”, or the three factors having the biggest impact on the project uncertainty on a Tornado plot. In this example, no data is available from previous field experience – hence, “engineering judgement” is applied. The NPV is in millions of US$.
As discussed above, it is necessary to recognise the uncertainty in our knowledge of the value of the various factors that are combined in the calculation of the project's Net Present Value. Each cost and performance element has a probability distribution curve associated with it. An accurate application requires considerable engineering judgement, but a normal, triangular or a Weibull (skewed) distribution can be used (figure 12). It is worth noting that, as a consequence of the shape of this type of curve; many cost estimates have a skewed distribution. This implies that the lowest possible cost is not much lower than the most likely cost, but the highest possible cost can be many times higher than this most likely cost. 0.025
Normal Distribution Weibull Distribution Trianglar Distribution
0.020
0.015
0.010
0.005
-30
0
30 NPV [million $]
60
90
120
10
SPE 82224
Figure 12: Three types of distributions suitable for engineering applications; Normal distribution, Weibull (skewed) distribution and a Triangular distribution.
3.3 Selection of Injection Water Source for the A2
different conclusion would have been reached if the full study process were repeated for the lower quality A2 reservoir. 4.
Reservoir The selection criteria for injection water source discussed above relates only to matrix injection into the high quality, A1 Reservoir. Reservoir A2, however, is relatively tight and studies have shown that fracturing is required to achieve and maintain the target injection rate. Thermally induced fractures have the potential of being contained within the target zone if the injection pressure is carefully controlled [11], [12]. Injection of cold water will result in big difference in temperature between the cold, water–flooded zone and the warmer in-situ formation. It is especially important that the injection water be clean during the initial injection well startup phase, when a matrix-like injection profile should be maintained prior to initiation of the thermal fracture, as this will ensure sustainable injection of cold fluid to cool the formation The injection water sources that can be considered are: aquifer water, relatively clean and cold, seawater, likely to cause serious scaling problems in the production wells and filtered and cooled produced water, commingled with aquifer water. These three options can now be compared in the same manner as discussed above for the high quality, A1 reservoir. The above process not only assigns a quantitative value to each volume of injection water, but also provides a quantitative framework against which day-to-day decisions concerning operation of the waterflood plant and process can be made.
4. Conclusions 1.
2.
3.
Water injection is an essential element in modern oil field operations, improving oil recovery and adding a value to the asset. Technical and economic aspects of water injection are of equal importance as the more traditional facets of oil field planning and operation. Application of this holistic overview leads to the development of a six–stage process tool for quantification of the value of injection water. Knowledge of this “Water Value” can be used to develop an optimum waterflood design. The process has been illustrated by its application to a North Sea type field case. A techno–economic evaluation of the most suitable injection water source for this North Sea type field concluded that injection of commingled aquifer- and produced water is the most attractive economically for the studied case of matrix injection into the high quality A1 reservoir. It is indicated why it is most likely that a
The above process not only assigns a quantitative value to each volume of injection water, but also provides a quantitative framework against which day-to-day decisions concerning operation of the waterflood plant and process can be made.
Acknowledgements The authors are pleased to recognise the contributions made by Decisioneering Inc., provider of the Crystal Ball® program. One of the authors is pleased to recognise the financial support from the sponsors of the Produced Water Reinjection Project and from Schlumberger.
Nomenclature AW BHIP ∆Ew NPV PW Qo Qw PRes SW t ∆W
Aquifer Water Bottom-Hole Injection Pressure [bar] Incremental value of injection water [$/bbl] Net Present Value [$] Produced Water Oil production rate [m3/day or bpd] Water injection rate [m3/day or bpd] Reservoir Pressure [bar] Sea Water Injection time [days] Incremental volume of Water injected [m3]
References 1.
2.
3. 4.
5.
6.
Todd, A.C., J.E. Somerville, and G. Scott. The Application of Depth of Formation Damage Measurements in Predicting Water Injectivity Decline. Paper SPE 12498, presented at the Formation Damage Control Symposium. Bakersfield, CA. 1984. Dunn, M.D.: A Method to Estimate the Value of Well Log Information. Paper SPE 24672 presented at the 67th Annual Technical Conference and Exhibition. Washington, DC. 1992. Lien, S.C. et al. Brage Field, Lessons Learned After 5 Years of Production. Paper SPE 50641, presented at the SPE European Petroleum Conference. The Hague. 1998. K. I. Andersen et al. Water Management in a Closed Loop - Problems and Solutions at Brage Field. Paper SPE 65162, presented at the SPE European Petroleum Conference. Paris. 2000. Bakke, S. et al. Produced Water Re-Injection (PWRI) Experiences from the Ula Field, In Produced Water 2. Proceedings from the 1995 International Seminar on Produced Water, Trondheim, Norway, 25-28 September. Plenum Press: New York. 1995. Hjelmas, T.A. et al. Produced Water Re-Injection: Experiences From Performance Measurements on Ula in the North Sea. Paper SPE 35874, presented at the SPE International Conference on Health, Safety and
SPE 82224
Environment. New Orleans. 1996. van der Zwaag, C. and Øyno, L. Comparison of Injectivity Prediction Models to Estimate Ula Field Injector Performance for Produced Water Reinjection. In Produced Water 2. Proceedings from the 1995 International Seminar on Produced Water, Trondheim, Norway, 25-28 September. Plenum Press: New York. 1995. 8. Sharma, M.M. et al. Injectivity Decline in Water-Injection Wells: An Offshore Gulf of Mexico Case Study. Paper SPE 60901, presented at the SPE Production & Facilities Conference, February 2000. 9. Paige, R.W., et al. Optimising Water Injection Performance. Paper SPE 29774, presented at the SPE Middle East Oil Show. Bahrain. 1995. 10. Bayona, H.J. A Review of Well Injectivity Performance in Saudi Arabia's Ghawar Field Seawater Injection Program. Paper SPE 25531, presented at the SPE Middle East Oil Technical Conference and Exhibition. Bahrain. 1993. 11. Stevens, D.G., L.R. Murray, and P.C. Shah. Predicting Multiple Thermal Fractures in Horizontal Injection Wells; Coupling of a Wellbore and a Reservoir Simulator. Paper SPE 59354, presented at the SPE/DOE Improved Oil Recovery Symposium. Tulsa, Oklahoma. 2000. 12. van den Hoek, P.J. and J.D. McLennan. Hydraulic Fracturing in Produced Water Reinjection. Paper presented at the Workshop on Three-Dimensional and Advanced Hydraulic Fracture Modeling, held in conjunction with the Fourth North American Rock Mechanics Symposium, Seattle, WA. 2000. 7.
11
