Documenti di Didattica
Documenti di Professioni
Documenti di Cultura
This paper was prepared for presentation at the International Petroleum Technology Conference held in Doha, Qatar, 79 December 2009.
This paper was selected for presentation by an IPTC Programme 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 International Petroleum Technology Conference and are subject to correction by the author(s). The material, as presented, does not necessarily
reflect any position of the International Petroleum Technology Conference, its officers, or members. Papers presented at IPTC are subject to publication review by Sponsor Society
Committees of IPTC. Electronic reproduction, distribution, or storage of any part of this paper for commercial purposes without the written consent of the International Petroleum Technology
Conference 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, IPTC, P.O. Box 833836, Richardson, TX 75083-3836, U.S.A., fax +1-972-952-9435.
the gap between technical understanding and the actual questions that must be asked about the interpretations,
field data during short- and long-term injection fall-off such as 1) What do we actually see into formation, i.e.
tests as they become more prevalent nowadays in the injected water or reservoir oil? 2) Do the permeability and
petroleum market place. skin values change with time? 3) How can we capture and
understand reservoir changes versus time?
Problem Setting
Simulation Model and Results
The evolution of testing tools has provided solutions of
new interpretation algorithms and methods. The A three-dimensional dynamic simulation has been
downhole pressure transient testing remains an important selected as a means to show the importance of the factors
phase in field exploration and production to deliver mentioned in the previous section, understand fall-off
reliable information about in-situ formation conditions. peculiarities, and discuss the way the outputs can be
One particular output from the well test analysis is the achieved when testing injection wells during continuous
formation permeability, i.e. a parameter critical to the pressure maintenance. The simulation has been set up and
determination of field production capacity and planning run using a dynamic simulator. The model was designed
of downhole operations for maximizing hydrocarbon to be a radial-grid with the grid size logarithmically
recovery. refined from the wellbore diameter of 12 inches and
outwards up to an external diameter of 85000 ft, for
When performing pressure transient tests involving a well graphical illustration, see Figure 1. Such configuration is
shut-in operation in a single-phase system, an answer to due to a typical radial propagation of the injection flow in
the permeability is straight forward, i.e. the value is the near-wellbore of a vertical well as well as to ensure
determined from the stabilized section of the pressure the wellbore pressure transient data recorded at a high
derivative line from the pressure derivative log-log sample rate is accurate and the boundaries are not felt
plot [2], and its magnitude is dictated by how high or low during a fall-off period. Another reason of using a
the stabilization section is with respect to the logarithmic refining approach is to exclude large number
dimensionless pressure in a general type-curve analysis of cells in the model that would severely slow down the
approach. This type of analysis has widely become calculation time. The input data used in the simulation are
computer-aided nowadays. However, in a two-phase stated in Table 1. For reservoir pressure maintenance
system, particlualrly for water injecting wells in a issues, a production well was considered in the radial
waterflooding development plan, the conventional model located at 6600 ft away from the injecting well
interpretation approach does not consider the mobility with the production flow rate equal to the injected rates
change between the injecting phase and the formation under complete voidage replacement consideration.
fluid. It becomes an important issue to consider the
volume and the associated injection radius occupied by To understand the effect of wettability system on the log-
water pumped into formation. The on-going injection log response, oil-water fluid model with three (3)
creates an individual pressure response with additional scenarios has been considered corresponding to different
stabilization plateau during fall-off and commonly it is wettability systems: water-wet, intermediate-wet, and oil-
quite apart from the one created by the formation fluid wet, respectively. The examples of the wettability systems
due to difference in effective permeability, relative used in the model are shown in Figure 2 (a), (b), and (c).
permeability, phase mobility and rock wettability issues. The relative permeability curves have been generated
The response becomes mored dramatic in the case when artificially using a dedicated software suite.
long-term injection operation is followed by a short fall-
off that does not have enough duration to see beyond the Each wettability system scenario has featured a post-perf
waterflooding front, i.e. the value of the investigation (short-term) injection period followed by a basic fall-off
radius is small. In such a case, the permeability and then with four additional cases corresponding to
determined will represent only the water phase. Should different durations of a main injection period as can be
the value be taken for the permeability due to oil, it will seen from Figure 3. The main injection period is then
result in overestimating the reservoir performance and followed by a final (long-term) fall-off. The durations of
production forecast. On the opposite side, when an the main injection period are equal to seven (7), thirty
injection fall-off is analyzed at the beginning of a well life (30), ninety (90), and one hundred and eighty (180) days,
after short post-perf injectivity, the radial flow respectively, and for observing the dynamics of a
stabilization plateau due to water can not be identified as transition stage between water-related and oil-related
it is masked by wellbore storage and only the one due to stabilization plateaus (phase mobility change). For the
oil is visible in the plot. At this early injection, the sake of simplicity, the injection rate was made constant
waterflooding front has not propagated deep into the thoughout the simulation and equal to 5000 bbl/d. The
reservoir yet. Therefore, an appropriate methodology total volume of the injected water its corresponding radius
should be devised and applied in test design and of injection has been calculated for each main injection
interpretation to deliver accurate commercially valuable period case and is stated in Tables 2 and 3. During short-
well test outputs, specifically for short-lived fall-offs. term and long-term fall-offs simulation runs have been set
Summarizing the relevance of the above, there are three to record bottomhole flowing pressure at every 5 secs.
IPTC 13440 3
The duration of both basic and final fall-offs were 24 hrs. where r inj is the radius of injection in ft; V inj is the
After a run is completed, fall-off pressure readings were
extracted and represented on a log-log comparison injected volume of water, bbl; h net is the net formation
pressure derivative plot.
thickness, ft; is the formation porosity; S wi is the
The results of simulation are shown in Figure 4. Cases (a), initial water saturation, and S orw is the residual oil
(b), and (c) are corresponded to the above-mentioned saturation, respectively. The last three values mentioned
wettability systems, respectively. Since the recording of are dimensionless in units.
fall-off pressure was simulated to be at the wellbore face,
the wellbore storage is not considered in the plots. In To understand the dynamics of saturation propagation
reality, the first 1.5 to 2 logarithmic cycles should be within the formation, a plot of water saturation
attributed to the wellbore storage domination period that
distorts the reservoir response. From the plots mentioned, distribution, S w , at various distances from injector well
two radial flow stabilization plateaus are clearly seen. The versus injection program duration has been created from
upper one corresponds to the radial flow due to water simulation results. A case for water-wet wettability
(marked as effective water permeability @ S orw ) and the system is shown in Figure 5 (a) and (b). For distribution
plot covering the first 20 days of injection, see
lower due to oil (marked as effective oil permeability Figure 5 (a), note the stabilized straight lines of water
@ S wi ), respectively. A step-down phase change (or saturation between Day 1 and Day 2. These are due to the
mobility change) as indicated by the inflection point well being shut-in for a basic fall-off that was considered
between two plateaus in the derivative plot, is also seen. in the simulation model after the short-term injectivity
This corresponds to the phase change between the water test. Here is an example to explain the meaning of this
and the formation fluid. In the subsequent fall-off events, distribution. For an example to examine the water
it propagates further as observed along the shut-in time saturation behaviour at 300 ft away from the injector, an
against the corresponding durations of the main injection observation is made at line 12, Figure 5 (b) that
period. The greater the volume of water injected during corresponds to a distance of 275 ft. This figure of 275 ft is
the main injection period, the further in time the phase taken as the closest arbitrary distance to the interest of this
change response in pressure is observed on the log-log investigation because the non-rounded distance values
plot. presented in Figure 5 (a) and (b) are the result of
summation of radial block sizes used in the simulation
Aslo of note, is that the position of radial flow plateaus model in a logarithmically refined manner. During the
with respect to each other and the shape of the phase first fifteen (15) days the water saturation in the formation
change section in the log-log plot are governed by the at the given distance remains equal to the initial water
type and signature of wettability system, i.e. relative saturation or water connate saturation since the volume of
permeability curves, as previously shown Figure 2 (a), injected water has yet to arrive. For our water-wet
(b), and (c). The greater the difference between oil and scenario, the initial water saturation, S wi , is equal 0.26
water end-point permeability values, the more their radial and corresponded to zero relative water permeability in
flow plateaus are apart from each other. The contrast Figure 2 (a), mentioned previously. As the water front
intensity of the phase transition zone depends on how arrives at the given distance, the water saturation starts to
expressive is the rock wettability to water phase. On the increase and the formation fluid gets desplaced by
contrary, for oil-wet rocks, the transition section is injected water until it is completely flushed. At this point
stretched along the logarithmic shut-in time range. For
intermediate-wet rocks having the characteristics of an the saturation reaches the residual oil to water, S orw ,
inflected oil-relative permeability curve (i.e., the end- equal to 0.25 in our case and the value corresponds to the
point relative permeability to oil shown in Figure 2 (b) is end-point water permeability in Figure 2 (a), respectively.
deliberately overstated for emphasis), the transition The water saturation recorded during radial phase change,
change in the log-log pressure derivative creates the hump i.e. within the range of S wi to S orw will correspond to a
in signature before it steps down to the radial flow due to
oil. value of relative permeability function, k rel , at the given
water saturation value S w . To get the absolute
For each case, the radius of injection has been calculated
using the formula shown in the below and the results are permeability, of formation, k abs , the product of the
stated in Table 3: following equation can be used:
k eff = k rel k abs ,
Vinj
rinj = 2.37 , where k eff is the effective permeability value that is
hnet (1 S wi S orw )
derived during downhole pressure testing analysis.
4 IPTC 13440
Another way to represent the saturation propagation in the the oil flowrate line overlaps with the region, where the
formation is given in Figure 6, where the water saturation water flowrate line reaches the highest magnitude. The
is plotted against the distance from the injecting well at latter confirms the fact the data about the oil phase is
various times. The distribution is less accurate to the one contained in the radial flow stabilization plateau due to
plotted previously in Figure 5 since with the increased water. In Figure 9, where the same propagation is
distance the size of grid blocks in the model gets larger represented against distance from the injector, one might
due to logarithmic refining and, hence, it decreases the see the wide extension of the oil flowrate. This explains
resolution accuracy. Like in Figure 5, the water saturation the reason of using the infinity-sized model, i.e. to
distribution vs. distance is provided for the water-wet avoid interaction from the model impermeable boundaries
scenario. that will reveal itself from the type-curve pressure
transient analysis in the log-log plot.
Thus, discrepancy in phase permeability values at water
injection fall-off tests are mainly determined by relative Field Background and Reservoir Challenges
permeability curves. The first radial flow stabilization
plateau on a pressure derivative log-log plot determines To test if the hypothesis can be observed in the field,
the effective permeability to water while the later one to injection well pressure data was obtained from a
oil, respectively. However, a question can be raised here deepwater oilfield in the Malay Basin, offshore Sabah
whether the radial flow stabilization plateau due to water (East Malaysia). Water injection is a key component of
could contain information about the oil phase. To provide the project, and required for pressure maintenance, sweep,
an answer and understand the effect of oil phase on the and to attain a high recovery factor. Permanent downhole
pressure derivative curve vs. elapsed time, the residual gauges (PDG) were installed in all water injection wells
phase-by-phase flowrate propagation within formation at on the field to allow for continuous pressure surveillance
different times, after the well was shut-in for fall-off, was without intervention. Both planned and un-planned
generated. For the sake of simplicity, only the basic fall- pressure fall-off data is recorded at high frequency, and is
off and the one following the one-week main injection used for pressure transient analysis. Planned fall-off
period were used in the analysis. For their responses on a periods include injectivity tests (usually initially on a
pressure derivative log-log plot, a series of three nodal well) and as part of regular field surveillance activities.
points were selected: at the beginning, in the middle, and Un-planned fall-offs occur occasionally when there are
at the end of a phase transition, see water and oil phase shut-downs in the injection system, and the wells shut
series 1-a, 1-b, 1-c and 2-a, 2-b, 2-c in Figure 7. Then, automatically.
contribution of phase-by-phase saturation related peak
flow rates radially propagated at the above mentioned The water injection well used as an example, is of subsea
points from the wellbore away into a reservoir during the wet tree design, on a manifold with several other injection
abovementioned fall-offs were derived. The association wells in water depths of around 1300m. The well was
with the pressure derivative curve to determine effective perforated in a single, homogenous and high-quality
phase permeability values is shown in the sequence of sandstone reservoir of 12m net thickness. The well angle
plots from Figure 7 to Figure 9. In Figure 8, the through the reservoir is 66 degrees, and the well is located
propagation is plotted against the model cell number and within the oil zone. Immediately post-perforation an
the injection well is at zero cell number (for our case with initial injectivity test and fall-off was performed on the
logarithmically refined model the cells get greater in the well. Subsequently, a visco-elastic gel (VES) frac
radial size away from the injector summing up to 85000 ft operation was carried out to improve the injectivity of the
as recorded in Table 1, so the refining effect explains the well. Continuous injection commenced 12 months later
distorted response of a phase flowrate at larger cell with no changes to well injectivity after the shut-in
numbers in Figure 8). period. Typical injection rates are around 24,000 bwpd.
It is seen from the figure there is not much of water phase Injectivity Analysis Challenges
front propagation into the reservoir within a fall-off (no
injection during the well shut-in), as observed in water One of the prime aims of injectivity tests is to verify
series 1 for the first fall-off and that of series 2 for the dynamic (pressure transient analysis-derived)
final fall-off, respectively. As the water stops propagating, permeability-thickness, kh , with that derived from
its flowrate magnitude attenuates with the elapsed fall-off petrophysical logs. In the case of water injection into an
time, see sub-indices a, b, and c within each oil zone, interpretation of the initial, early injection and
water series 1 or 2, see their corresponding locations in fall-off periods can be problematic unless the presence of
the plots, i.e. from Figures 8 and 9. On the contrary, the the water flood-front is recognized and accounted for in
oil phase continues propagating away from the injector the pressure transient analysis. In addition to this, the
with less attenuation compared to that for the water and effect of temperature on injected fluid viscosity is
there is domination in ratio of the oil flowrate over the important as this will vary the mobility ratio between the
water one, respectively. Besides that, the phases are water and oil. Over time and continued injection, cold
overlapping with respect to each other. For the nodal injected water will warm as the flood front progresses into
point at the beginning of the phase transition; see series 1,
IPTC 13440 5
the reservoir, and the mobility ratio between the fluids designed to include a series of step rates to evaluate the
will change. injectivity capability and to feature a long stabilized
injection period to ensure the injectivity is stabilized.
During several well injectivity conformance and fall-off However, after few step rate series, the injection pressure
test technical reviews, it is found that there can be a increased above the estimated fracture initiation pressure
significant discrepancy between the initial well-test and no significant sign of fracturing was observed. The
derived permeability-thickness, kh resulted from post- injection rate was reduced, and it was expected that the
perf short-term injectivity and that from the petrophysical pressure would decrease as well. However, the pressure
log-derived. This discrepancy can be as high as multiple kept increasing, so the injectivity test was halted, with a
of five (5). At the same time, the long-term planned total of around 1300 bbl of water injected. Following this,
injectivity which was carried out after one year of the well was shut in for pressure fall-off for less than
continuous water injection, shows relatively good 2 minutes. The pressure transient data have shown to be
coherence of derived kh result compared to that from the dominated by wellbore storage period and no radial flow
log-derived petrophysical analysis. This has created a was observed, as shown in Figures 11 and 12.
sense that the interpreted perm-thickness result from the
post-perf short-term well-test analysis is dubious until (2) Short-term injectivity test during VES fracturing
proven otherwise. campaign
With the objectives of the post-perforation injectivity test
From the actual field data, as shown in Figure 10 [3] shows not achieved, one month later, Well A was brought back
viscosity versus temperature for this particular formation. from suspension for the second injectivity test as a part of
The viscosity of water at the reservoir temperature is the the VES fracturing operation [3]. This time, the injection
same as the oil viscosity (0.3 cp). If water is injected at and fall-off tests consisted of a series of five injection and
cooler temperatures than the ambient reservoir fall-off sequences. All fall-off tests lasting from one to
temperature, it may in turn cool the reservoir fluids near three hours proved to have the same radial flow
the well bore, thereby increasing the viscosities of these stabilization plateau, even though the skin value changed
fluids. The viscosity difference is less than three times of at every fall-off due to the VES fracturing events. The
the oil viscosity at the near wellbore region and then radial flow stabilization plateau delivered the permeability
temperature increases again away from the wellbore. This value of 350 mD and was analysed as oil effective
thermal effect is insufficient to explain why the permeability since the total injected volume for this
permeability derived from short injectivity analysis was operation was less than 3500 bbl. This left the radial flow
observed to be significantly higher than the long term due to water unidentified again due to the wellbore
injection data. To investigate the causes for the storage period domination in the pressure derivative log-
discrepancies, a 3-D dynamic simulation to generate a log analysis plot (as shown in Figures 13 and 14).
representative reservoir injection pressure response while
the injection front propagates was built. The transient (3) Main injection program with pressure recorded
analysis result from this model-derived pressure with PDG
derivative is then verified and compared with analysis The continuous injection program commenced twelve
from actual field data. With an encouraging match months later after the VES fracturing operation was
between these two transient analyses the study has been completed. Data analyzed for this paper was recorded
carried forward, with some recommendation of further over eight months of continuous injection with an average
studies in the future. water injection rate of 24 000 bbl/d. The well was
equipped with PDGs and the bottomhole pressure data
In the following discussion, field examples of the were captured in real time for reservoir monitoring
previously discussed pressure transient analysis impact purposes. Fall-off data were obtained when the well was
from early-time, middle time until the late-time injection shut-in for operational purposes. The injection pressure-
events are highlighted. From the post-perforation (i.e. rate history from PDG is shown in Figure 15. For selected
early-time) injectivity test, to the near wellbore treatment fall-off periods marked with light blue color in the figure,
(i.e. middle-time) injectivity events until the long-term a comparison pressure derivative log-log plot was created
(i.e. late-time) planned injectivity test performed, the and shown in Figure 16. The plot also captures the fall-off
observation of the injectivity results is made with respect from the post-perforation injectivity test, Fall-off #1, and
to the topic of this paper. the three most representative fall-off data taken from the
VES fracturing period, Fall-offs #2 to #4, respectively.
(1) Post-perforation injectivity test The rest of the data (i.e. from Fall-off # 5 to Fall-off #10)
This injectivity test of Well A was conducted right after correspond to the main injection period. On the whole, the
the formation was perforated and lasted around 2 hours. pressure fall-off data of the Well A follows our
The test objective was to establish injection assurance by hypothesis: two stabilization plateaus to water and oil are
pumping treated seawater at maximum stabilized rate of seen with steady propagation of the waterflooding front
25000 bbl/d or reaching the wellhead pressure of with distance and in time. This radial flow stabilization
3,000 psi, whichever occurs first. The injectivity test was plateau corresponding to the effective water permeability
at the residual oil saturation is almost five times less than
6 IPTC 13440
the one corresponding to the effective oil permeability at incorrect data might have impact on decisions about well
the initial water saturation. In the figure, where the fall- design, completion, production planning and forecasting,
off derivative is updated with the input of correct rate especially if the data was acquired during the appraisal
history, fall-offs #1 to #5 show that only the radial flow stage of a field. Consequently, it is important to design an
stabilization plateau due to oil is observed. Fall-off appropriate pressure transient testing program in order to
periods #11 and #12 are too short, such that the injection allow identification of both of the radial flows in one fall-
front transition is way beyond the total investigation off period. This idea of a proper injectivity testing
radius. Fall-off #14 delivers a good radial flow program in order to fulfill the testing objectives
stabilization plateau due to water as its radius of highlighted in this paper will be discussed with
investigation is greater than the radius of injection (i.e. the recommendations in the next section.
injected volume radius). In general, the comparison
pressure derivative log-log plot shows the significance of The study demonstrates the needs to understand the
integrated illustration of all pressure fall-off data. limitation of pressure transient analysis, particularly with
computation of the radius of investigation that is based on
Table 4 lists the results obtained from fall-offs including the assumption of perfect cylindrical radial displacement
the effective permeability to water, the skin, and the due to homogeneous formation parameters. In reality,
mobility ratio. It can be seen that the average permeability there will be forces acting to distort the flood front away
remains around 113 md. The high skin value at the from cylindrical, i.e. due to gravity, geological
beginning corresponded to the fracturing period with heterogeneity, dip of formation, and lateral geological
visco-elastic gel and breaker injections. As the main changes in all directions away from the well. The real
injection program commenced, skin became negative and data will thus have a smeared transition between oil and
steady at values of -2.5 to -3.0 as a result of maintaining water phases, with the shape of the transition roughly
high injection rates at 24 000 bbl/d. Therefore, it can be corresponding to a fractional water curve in a vertical
said that skin can change with time at the beginning as the sense. Areally, with the dip and thickness changes with
result of post-perforation operations, cleaning up, distance away from the well, and in the example case a
formation treatment and fracturing. However, in contact downdip, the true transition shape is likely to
consideration of selecting the radial flow and keeping remain unknown and it will not be a perfect cylindrical
flow rates unchanged, the final variations of skin are shape. In some sequences of later fall-offs (#7, #9, and
reasonable. The last column, the mobility ratio, shows the #13 as examples in Figure 16), one cannot observe two
ratio of mobility in the inner zone (water) to mobility in stabilization plateaus but the pressure responded to a
the outer zone (oil). For fall-offs #2, #4, and especially flattening or thinning of the water displacement front that
#5, where two radial flows are identified, it almost could mean a higher actual value of the investigation
corresponds to the ratio of end-point phase relative radius.
permeability values. As the injection program continues,
the ratio approaches unity. This could be due to the The injection radius data obtained from analysis should be
following reasons: (a) phase transition getting smeared compared with that from the 3-D dynamic field model,
due to geological heterogeneities, (b) actual reservoir which incorporates the actual well injection history. The
thinning away from the wellbore, and (c) change of water modelled radial displacement can show sagging effects
viscosity due to thermal effects. like coning of the pumped water near the wellbore and
when the waterflooding front propagates deeper into
The importance of selecting the proper radial flow reservoir due to gravity effects and formation thickness
stabilization plateau and the effect of the application of changes away from the wellbore.
the output permeability-thickness for specific reservoir
engineering purposes, can be illustrated by examining the Test Design and Interpretation Considerations
fall-off period #5, which is at an early stage of the water
injection program, as shown in Figure 17. The radial flow Problems stated in the previous sections and observations
due to water is not seen here as it is masked under the from field data have raised a reasonable issue on the
wellbore storage effect. Selecting the visible radial flow needs for fall-off test design to recognize both radial flow
due stabilization plateau to oil, Figure 17 (a), and regimes and the phase transition, respectively. It also
interpreting it using water parameters delivers the becomes clear that observing the waterflooding front
overstated results for permeability and skin, i.e. away from a water injection well is generally only
k = 520 md and S = +15.7, respectively. However, possible early in the field life. Therefore, to gather the
putting the radial flow up to a level, where the required information, a series of fall-offs should be
stabilization plateau due to water is, Figure 17 (b), as incorporated at earlier stage of the water injection plan.
observed from later fall-offs, where an inflection point How often these fall-offs should be run and how long they
exists in between two stabilization plateaus, the results should be is much in relation with the formation
become k = 114 md and S = -1.0, respectively. The permeability, which again is the main uncertainty.
decrease between the permeability results is more than Therefore, it is advisable to run petrophysical log to get
4.5 times. There are obvious consequences that the the near-wellbore permeability value and use it as the
initial parameter for subsequent fall-off designs.
IPTC 13440 7
radius to be at least five times greater, i.e. r inv =5 r inj . At all stages of planning, a combination of computer-
Then, the r inv value can be recalculated to know how aided well test design, single-well and full 3-D dynamic
reservoir models should be used including sensitivity
long a fall-off should be in duration to cover the analysis for a reasonable range of parameters.
abovementioned events. Collaboration with geologic and petrophysical disciplines
is a mandatory to identify and verify those scenarios that
Alternatively, when analyzing acquired data, if the r inj might look as controversial at the beginning.
can be estimated from actual pressure data plotted on a
log-log pressure derivative graph, where the inflection Conclusion and Recommendations
point in the derivative line exists and shows two
stabilization plateaus, then rough comparison with the 1. This paper presents examples of the challenges in
analyzing fall-off tests in the two phase injection
r inj value based on perfect cylindrical displacement system. Information about wettability and relative
equation can provide some qualitative insight into flood permeability is required in order to obtain the correct
front heterogeneity. interpretation. However, this data is not always
available. As a result, an integrated study of all
If the expected inflection point is not observed in the log- transient data with corrected rate history is required
log pressure derivative line, instead only one stabilization for pressure transient analysis.
plateau to water is seen, it could due to causes, such as (a) 2. A three-dimensional dynamic model was built to
water injection-front flattening effect, (b) overall reservoir simulate the effect of injecting water into oil
thinning away from the wellbore, or (c) change of water reservoir for oil-wet, intermediate wet and water-wet
viscosity due to thermal effect. As for the flattening of the systems.
water front due to gravity sagging, geologic 3. Radially propagated waterflooding fronts as observed
heterogeneities, as well as lateral formation thickness from the model were used to understand the changes
changes, the investigation and comparison of the injection in log-log pressure derivative stabilization plateaus,
radius values is advisable to carry out against 3-D and lead to correct interpretation.
dynamic field modeling that would incorporate fall-off 4. Field data were then analysed to corroborate the
scenarios identical to as-run ones. As for the water discussed hypothesis, in order to determine whether
viscosity issue, it is related to mobility and diffusivity real-world uncertainties hinder the observation of the
components attributing to composite zone behaviour. In waterflooding front in the pressure transient analysis.
theory, the mobility ratio is dictated by a ratio of relative 5. Care must be taken when analyzing fall-off tests from
end-point phase permeability values. However, in reality, injectivity tests into the oil zone in the early life of an
when the cold water is pumpued down into formation, it injection well, or during field appraisal. Incorrect
changes its viscosity as the result of the water warming up calculation of absolute permeability and skin could
while propagating away from the wellbore. The change lead to poor development and design decisions.
could be such, so both the mobility and diffusivity ratios
will remain close or equal unity. This effect can also be References
seen in our case, see later fall-offs #9 and #13 in
Figure 16. For these fall-offs, the transition period is no 1. Bourdet, D.: Well Test Analysis: The Use of
longer observed in full (Fall-offs #10, 11, 12, and 14 are Advanced Interpretation Models, Handbook of
too short to make a proper conclusion). In such a case, Petroleum Exploration and Production 3, Elsevier,
additional considerations should be taken to see the water Netherlands (2002).
thermal effect on the success of pressure transient test. 2. Bourdet, D., Whittle, T. M., Douglas, A. A., and
Pirard, Y. M.: A New Set of Type Curves
With both the radial flow plateaus clearly identified from Simplified Well Test Analysis, World Oil (May
log-log plot, the value of performing diagnostics comes 1983), pp 187-195.
8 IPTC 13440
3. Webb, T., Omar, J., Daungkaew, S., Lim, L. M., presented at the 2009 SPE European Formation
Tibbles, R., Munoz, I., Morales, H.: Overcoming Damage Conference held in Scheveningen, The
Poroelastic Behaviour in the Water Injector Wells in Netherlands, 2729 May 2009.
the Kikeh Deepwater Development, SPE 122018
Water-Wet Relative Permeability Function Intermediate-Wet Relative Oil-Wet Relative Permeability Function
Permeability Function
1.0 1.0 1.0
0.9 0.9 0.9
Relative Phase Permeability
Post-Perf Injectivity Test with Fall-Off Main Injection Period Final Fall-Off
10000
Qinj, bbl/day
Log-Log Comparison Plot for Oil-Wet Case, dp and dp' normalized [psi] vs. dt
Table 2 Total volume of water injected
10000
Pressure [psia]
After 30 days 155 000 bbl
After 90 days 455 000 bbl 100
Total mobility Effective water permeability @ Sorw
Water Intermediate Oil After 1 day After 1 week After 1 month After 3 months After 6 months
0.8
Total mobility Effective water permeability @ Sorw
1
2
0.7
3
1000 4
Water Saturation
0.6
Pressure [psia]
0.5 5
100 6 7 8 9 10 11
0.3
10 0
0.001 0.01 0.1 1 10 100 0 2 4 6 8 10 12 14 16 18 20
Time [hr] Time [days]
After 1 day After 1 week After 1 month After 3 months After 6 months
(a)
(a)
Water Saturation Distribution in Time
Log-Log Comparison Plot for Intermediate-Wet Case, dp and dp' normalized [psi] vs. dt
1
10000 Legend:
10 - @ 168 ft; 11 - @ 215 ft; 23 - @ 275 ft; 13 - @ 351 ft; 14 - @ 449 ft;
0.9
15 - @ 574 ft; 16 - @ 733 ft; 17 - @ 935 ft; 18 - @ 1195 ft
0.8
Total mobility Effective water permeability @ Sorw
0.7
1000
Water Saturation
0.6
Pressure [psia]
0.5
10 11 12 13 14 15
0.4
16
100
0.3
17
18
Total mobility Effective oil permeability @ Swi
0.2
Total mobility Effective water permeability @ Sorw
Total mobility Effective oil permeability @ Swi 0.1
10 0
0.001 0.01 0.1 1 10 100 0 20 40 60 80 100 120 140 160 180 200
Time [hr] Time [days]
After 1 day After 1 week After 1 month After 3 months After 6 months
(b)
(b)
Figure 5 Distribution of water saturation in formation at various
distances in time, water-wet scenario: (a) within first 20 days; (b)
along the whole water injection program of 180 days
10 IPTC 13440
1.0
0.9
0.8
Total mobility Effective water permeability @ Sorw
0.7
Water Saturation
0.6
0.5
0.4
0.3
0.1
0.0
0 100 200 300 400 500 600 700 800 900 1000
Distance [ft]
1.4
Based on 35,000 ppm seawater @ 4800 psi
1.3
Log-Log Comparison Plot for Water-Wet Case, dp and dp' normalized [psi] vs. dt -1.031
y = 81.711x
1.2
10000
1.1
Viscosity, cP
0.8
reservoir
0.7
temperature
1000 of 188 F
0.6 w = 0.80 1.16 cP
@ modeled injection
Pressure [psia]
0.5 temperature of 60 90 F
0.4
w = 0.68 cP @ test o = 0.32 cp @ reservoir
1-a 2-a 0.3 injection temperature of
temperature of 188 F
106 F (Well A)
100 0.2
0.0
50 60 70 80 90 100 110 120 130 140 150 160 170 180 190 200 210
10
0.001 0.01 0.1 1 10 100
Time [hr]
Figure 10 Comparison of reservoir oil viscosity, and injected water
After 1 day After 1 week
viscosity.
Figure 14 Log-log pressure and derivative plot of all fall-off periods kh = 20800 md-ft
k = 520 md
during and after VES frac operation Skin = 15.7
(a)
kh = 4560 md-ft
Figure 15 Injection pressure-rate history obtained from downhole k = 114 md
Skin = -1.0
permanent gauges
(b)
Figure 17 Fall-off #5 showing importance of selecting the proper
radial flow: (a) to oil; (b) to water