Exploration Well YC-1

Two sandstone oil reservoir intervals of the vertical YC-1 exploration well, formations L and U, were tested separately during the period from November 8th through November 16th, 2018. 

Formation L Well Test (DST 1) 

The first drill stem test (DST1) targeted the deeper formation L interval which is separated from formation U by an impermeable layer. 

Unfortunately the test had to be terminated sooner than expected due to the production of H2S during the initial clean-up. The well was shut-in for about 12 hours and then killed.

Table 1 shows the PVT and static data for this formation:

————————————–
Fluid Volume-Factor :1.3000 vol/vol
Fluid Viscosity :0.600E+00 CP
Fluid Compressibility :0.600E-05 1/PSI
Water Compressibility :0.300E-05 1/PSI
Porosity :22.3000 %
Water Saturation :20.0000 %
Net Thickness :150.000 FEET
Rock Compressibility :0.300E-05 1/PSI
Wellbore Radius :0.354E+00 FEET
Total Compressibility :0.8400E-05 1/PSI
—————————————

The DST1 test sequence is shown below:

(click on the plot to enlarge it)

The figure below shows the derivative of the PBU test. 

Time limit: 0

Quiz-summary

0 of 8 questions completed

Questions:

1. 1
2. 2
3. 3
4. 4
5. 5
6. 6
7. 7
8. 8

Information

get ready

You have already completed the quiz before. Hence you can not start it again.

Quiz is loading...

You must sign in or sign up to start the quiz.

You have to finish following quiz, to start this quiz:

Results

Time has elapsed

You have reached 0 of 0 points, (0)

Categories

1. Not categorized 0%

1. 1
2. 2
3. 3
4. 4
5. 5
6. 6
7. 7
8. 8

1. Answered
2. Review

1. Question 1 of 8

   1. Question

   What well and reservoir models could we identify from this log-log plot? (multiple choice)

   • Horizontal well in a homogeneous reservoir of infinite extent
   • Vertical well in a homogeneous reservoir with a sealing fault.
   • Vertical well in a heterogeneous reservoir of infinite extent.
   • Vertical well in a heterogeneous reservoir with a sealing fault.
   Correct
   

   -Vertical well with wellbore storage and skin in a homogeneous reservoir with a sealing fault.
   -Vertical well with wellbore storage and skin in a heterogeneous reservoir of infinite extent.

   Incorrect
   
2. Question 2 of 8

   2. Question

    

   Based on the static data in table 1, what is the permeability value and radius of investigation?
   hint:
   
   with Rinv: the radius of investigation in feet, K: the permeability in mD, ϕ: the porosity, µ: the viscosity in cp, Ct: the total compressibility in 1/psi and ∆t: the elapsed time in hours.
   

   • K= 36.7mD and Rinv ~ 410 ft
   • K= 36.7mD and Rinv ~ 630 ft
   • K= 73.4mD and Rinv ~ 896 ft
   • K= 73.4mD and Rinv ~ 731 ft
   Correct
   

   KH= 5,506.8 mD.ft, K= 36.7mD
   Rinv ~ 409 ft

   Incorrect
   

   As mentioned by Earlougher, R.C in the SPE monograph “Advances in Well Test Analysis” and Lee, J. in the “Well Testing” SPE textbook, the radius of investigation only applies to the infinite-acting radial flow case. With presence of boundaries/heterogeneities, the formula will not give a correct value. This is also consistent with SPE 102483 “The Use of Well Testing for Evaluation of Connected Reservoir Volume” by Michael Levitan, Mike Ward, Jean-Luc Boutaud de la Combe and Mike Wilson.
3. Question 3 of 8

   3. Question

    

   Could we use Deconvolution to improve our model identification?

   • Nope, Deconvolution won't help due to the short production time.
   • Nope, we only have 1 PBU test.
   • Yes, we could use the PBU test with an estimate of initial pressure from the MDT.
   • Yes, we could use Deconvolution once the conventional derivative is matched.
   Correct
   

   We need to use at least 2 PBU tests to refine the initial pressure and recover a reliable deconvolution response.

   Incorrect
   

   A few change in initial pressure could affect the shape of the Deconvolution response, in particular at late times. This is why it is recommended to use at least 2 PBU tests to refine the initial pressure and ensure the validity of the Deconvolution response.
4. Question 4 of 8

   4. Question

    

   What are two methods that we could use to obtain a minimum connected volume for DST1? Use one of these two methods to calculate a minimum connected oil volume at surface. (hint: bbl =ft³ x 0.17811 and ft³ =m³ x 35.31)
   

   • Using the unit-slope straight line on Deconvolution or matching the data with a closed reservoir model. Vmin= 4.3 mm stb of oil at surface
   • Using the unit-slope straight line on Deconvolution or using material balance with the flared volume. Vmin= 2.5 mm stb of oil at surface
   • using an analytical simulation or the total compressibility formula V= dV / (Ct dP). Vmin= 3.4 mm stb of oil at surface
   • using the radius of investigation or matching the conventional derivative with a closed reservoir model. Vmin= 1.9 mm stb of oil at surface
   Correct
   

   Closing the interpretation model using the conventional derivative or using the radius of investigation for a low side of the range. Vmin= 1.9 mm stb of oil at surface

   Incorrect
   
5. Question 5 of 8

   5. Question

    

   We obtained a “P-star”: P* = 5,165.8 psi when using the radial flow straight line on the Superposition plot.

   

   Could we use the P* value as the initial pressure for this analysis?

   • Yes, P*= Pi in the superposition plot.
   • No, this case shows that P* shouldn't be used to obtain the initial pressure Pi.
   Correct
   

   P* = Pi with an infinite acting reservoir. In this case, the late-time increase in the derivative suggests the presence of a boundary.

   Incorrect
   

   P* = Pi with an infinite acting reservoir. In this case, the late-time increase in the derivative suggests the presence of a boundary.
6. Question 6 of 8

   6. Question

    

   The plots below show the well test analysis and calculated parameters using a vertical well model in a closed homogeneous reservoir.

    

   Derivative plot

   

   Production History Plot

   

   Do you agree with the PBU analysis above?

   • Yes. All the plots are well matched.
   • The superposition plot is missing, this will show that the initial pressure is not correct.
   • No, there are some ondulations in the derivative at late times.
   • There is no need for a closed reservoir.
   Correct
   

   Incorrect
   
7. Question 7 of 8

   7. Question

    

   In the production history plot above, the model doesn’t match the first two drawdown periods. What could this discrepancy suggest?

   • The rate is not correct at the production start-up
   • The skin in the model is too low at the production start-up. The well is cleaning up.
   • The interpretation model is not correct.
   • A non-Darcy skin (turbulence) is missing in the interpretation model.
   Correct
   

   Incorrect
   
8. Question 8 of 8

   8. Question

    

   What would be the next steps in the well test analysis workflow (multiple choice)?
   

   • Select another interpretation model (open reservoir) and match the data.
   • Integrate the results with the other sources of subsurface information.
   • Find other solutions that match the data.
   • Write a beautiful report based on the solution above.
   Correct
   

   Incorrect
   

Formation U Well Test (DST 2)

After isolating the formation L, the operations proceeded to only test the formation U. The second drill stem test targeted formation U and its two layers U1 and U2. A Distributed Temperature Array was available inside the tubing-conveyed perforating TCP guns so as to measure the temperature profile along the two layers.  

The objectives of DST 2 were to determine the formation KH and skin factor, obtain some samples of the reservoir fluid at reservoir conditions and prove a connected volume of 10 million barrels of oil at surface. 

Table 2 shows the PVT and static data for formation U:

————————————–
Fluid Volume-Factor :1.3000 vol/vol
Fluid Viscosity :0.600E+00 CP
Fluid Compressibility :0.600E-05 1/PSI
Water Compressibility :0.300E-05 1/PSI
Porosity :22.3000 %
Water Saturation :20.0000 %
Net Thickness :150.000 FEET
Rock Compressibility :0.300E-05 1/PSI
Wellbore Radius :0.354E+00 FEET
Total Compressibility :0.8400E-05 1/PSI
—————————————

The MDT pressure correlation for the Formation U interval of YC-1 well is presented below.  

Below the downhole shut-in tool, the DST string included some downhole samplers and two gauge carriers at 3,315.5 mTVDss and 3,335.7 mTVDss, with four wireless pressure gauges per carrier. Oil rates were measured with a Vortex meter, gas rates with an orifice plate at the test separator. 

Both layers U1 and U2 were perforated and tested together. 

The pressure data for the test sequence as recorded by the downhole gauge 2812 is shown in the figure below. (click on the plot to enlarge it)

The clean-up flow was interrupted by an Emergency Shut Down (ESD) at surface. Once the clean-up was complete, the well was shut-in at the downhole tester valve for PBU 1. This was followed by a fluid sampling flow and a main flow period. The well was then shut-in downhole for the final PBU 2. 

No rate measurement was performed before the ESD and the choke model was used to estimate the oil rate during that period. During the test, the GOR increased from 650-680 to 1040 scf/bbl. According to the crew, the well could have produced below bubble point pressure. However, a single phase oil analysis was attempted in the section below. 

No production of H2S was observed.

The plots below show a comparison study between gauge 19420 (carrier at 3,315.5 mTVDss), gauge 2808 (carrier at 3,335.7 mTVDss) and gauge 2812 (carrier at 3,335.7 mTVDss).

Time limit: 0

Quiz-summary

0 of 15 questions completed

Questions:

1. 1
2. 2
3. 3
4. 4
5. 5
6. 6
7. 7
8. 8
9. 9
10. 10
11. 11
12. 12
13. 13
14. 14
15. 15

Information

get ready

You have already completed the quiz before. Hence you can not start it again.

Quiz is loading...

You must sign in or sign up to start the quiz.

You have to finish following quiz, to start this quiz:

Results

Time has elapsed

You have reached 0 of 0 points, (0)

Categories

1. Not categorized 0%

1. 1
2. 2
3. 3
4. 4
5. 5
6. 6
7. 7
8. 8
9. 9
10. 10
11. 11
12. 12
13. 13
14. 14
15. 15

1. Answered
2. Review

1. Question 1 of 15

   1. Question

   Why did the team install multiple gauges in the tubing? (multiple choice)

   • for redundancy (backup)
   • to remove the tidal effects
   • to monitor the fluid density between the gauges
   • to help recover a Deconvolution response
   Correct
   

   P* = Pi with an infinite acting reservoir. In this case, the late-time increase in the derivative suggests the presence of a boundary.

   Incorrect
   

   P* = Pi with an infinite acting reservoir. In this case, the late-time increase in the derivative suggests the presence of a boundary.
2. Question 2 of 15

   2. Question

   Could we extract some information about the fluid density between the gauges?

   • The pressure gradient is about 0.31 psi/ft. This gives 0.72 g/cm3
   • The pressure gradient is about 1.1 psi/ft. This gives 2.54 g/cm3
   • The pressure gradient is about 0.26 psi/ft. This gives 0.6 g/cm3
   • The pressure gradient is about 0.411 psi/ft. This gives 0.949 g/cm3
   Correct
   

   Incorrect
   
3. Question 3 of 15

   3. Question

   Which gauge would you select for well test analysis?

   • 19420
   • 2812
   • 2808
   Correct
   

   Incorrect
   

   2808 seems affected by some noise.
4. Question 4 of 15

   4. Question

    

   At the end of PBU 1, the downhole valve was opened up to prepare for the low flow rate sampling period. The reservoir engineer and the gauge engineers were confirming the time when the bottom-hole samplers would fire when they noticed a small pressure fall in the bottom-hole pressure gauges below the samplers. After checking more carefully, they noted that the fluid gradient between the downhole samplers had also increased to around 1.47 psi/m. During the sampling flow period, this fluid gradient stayed the same. The reservoir engineer adjusted the bottom-hole sampler timings and only took two samples during the low rate sampling flow. She reprogrammed the remaining samplers to fire at the start of PBU 2.

   Could you think of a reason why this gradient could have increased?

   • There could be some water falling down the well.
   • The presence of gas in the tubing could decrease the DP.
   • This could be due to gas expansion.
   • The team probably injected some methanol during shut-in.
   Correct
   

   It looks like the fluid could be water from the gradient. It could be that the well was flowing some water and at shut-in, the water still in the string fell down onto the tester valve. Once the tester valve was opened, it would fall into the well. Flowing the well at very low rates, the oil can bubble through the water and not lift it so the continuous phase is water and that is what a sampler would capture.

   Incorrect
   
5. Question 5 of 15

   5. Question

    

   The PBU data are shown in the log-log plot below:

   

   Why do we see some differences in the PBU responses, in particular at early and middle times? (multiple choice)

   • The permeability-thickness KH is increasing over time.
   • The ESD response shows a different derivative stabilization, probably due to some wellbore phase redistribution or rate uncertainty.
   • The ESD response shows a larger wellbore storage because of the surface shut-in.
   • We can note a decrease in skin from PBU 1 to 2.
   Correct
   

   ESD response shows a larger wellbore storage (surface shut-in) and a different derivative stabilization (issue with wellbore phase redistribution or rate uncertainty).
   We can also note an increase in skin from PBU 1 to 2.

   Incorrect
   
6. Question 6 of 15

   6. Question

   Is the measured GOR increase supported by the PBU data and other evidence?
   

   • Yes, we can see a clear increase in skin between PBUs 1 and 2.
   • Yes, the increase in wellbore storage suggests an increase in total compressibility, due to the increased gas concentration in the tubing.
   • The derivative shape with features 1 and 2 could suggest a multiphase region around the well at a lower effective permeability to oil. This would support the increase in GOR.
   • The derivatives are very consistent and do not change over time despite a supposed increase in GOR. In fact, the increase in GOR is likely an issue with the orifice plate meter. The DP across the gauges doesn’t change and is consistent with the MDT results.
   Correct
   

   The derivative shape with features 1 and 2 could suggest some issue with wellbore phase redistribution (increase in total compressibility) or a multiphase region around the well at lower effective permeability to oil. However, these features are very consistent and do not change over time despite a supposed increase in GOR. In fact, the increase in GOR is likely an issue with the orifice plate meter. The DP across the gauges doesn’t change and is consistent with MDT results.

   Incorrect
   
7. Question 7 of 15

   7. Question

    

   PBU 1 and 2 tests are shown in the superposition plot below.

   

   P* decreased from 4924.2 to 4904.7 psi between the two PBU tests. Could this decrease in pressure be interpreted as depletion?
   

   • This decrease could indeed be interpreted as depletion, with a small connected volume.
   • No, P* doesn’t have any meaning here, the reservoir is not of infinite acting.
   Correct
   

   We could also see from Deconvolution later on that pseudo steady state hasn’t been reached at the end of the test (the reservoir is still open).

   Incorrect
   
8. Question 8 of 15

   8. Question

    

   An attempt was made to recover a reliable deconvolution response over 128 hours, with two possible options below.

   a) Initial pressure Pi= 4,924.2 psi at gauge depth

   

   b) Initial pressure Pi= 4,921.4 psi at gauge depth

   

   Which Deconvolution and initial pressure would you select?

   • Deconvolution b) with an initial pressure Pi= 4921.4 psi
   • Deconvolution a) with an initial pressure Pi= 4924.2 psi
   Correct
   

   4921.4 psi- because the two deconvolution responses are consistent and from the notes, more build-up data are used from PBU 1.

   Incorrect
   
9. Question 9 of 15

   9. Question

    

   What is the minimum connected oil volume at surface, proven by DST 2?
   (hint: bbl =ft³ x 0.17811)

   • 21.6 mmstb of oil at surface
   • 31.6 mmstb of oil at surface
   • 18.9 mmstb of oil at surface
   • 25.4 mmstb of oil at surface
   Correct
   

   pore volume x 0.178 x (1-Sw) / Bo

   Incorrect
   
10. Question 10 of 15

    10. Question

     

    The figure below compares the conventional and deconvolved derivatives:

    

    Why are the conventional and deconvolved derivatives different at late times?
    

    • This is a distortion due to the way the conventional derivative is calculated.
    • Deconvolution is not valid
    • This is due to a short production time and long shut-in period.
    • This is because of multiphase flow in the reservoir (high GOR)
    Correct
    

    Incorrect
    
11. Question 11 of 15

    11. Question

     

    The reservoir engineer offshore suggested a problem with limited entry (limited perforations or partial penetration), identifying the derivative stabilization 1 as the radial flow regime across some limited height. If so, what would be the penetration ratio?
    

    • ~ 50%
    • ~25%
    • ~75%
    • ~7.95%
    Correct
    

    Incorrect
    
12. Question 12 of 15

    12. Question

     

    What other well and reservoir single-layer models could we investigate? (multiple choice)

    • horizontal well with the presence of boundaries
    • heterogeneous reservoir with the presence of boundaries
    • heterogeneous reservoir of infinite extent
    • dual porosity reservoir of infinite extent
    Correct
    

    The Deconvolved derivative shape could be indicative of some heterogeneities and/or boundaries.
    1: radial flow, 2: increase in mobility/storativity, 3: boundaries or decrease in mobility/storativity.

    Incorrect
    
13. Question 13 of 15

    13. Question

     

    If we assume that a reservoir crossflow is visible in the derivative plot and that both layers have similar permeability-thickness KH, could you comment on the skin distribution across the perforations? What could the derivative stabilizations 1, 2 and 3 represent? Could we extract the permeability in layers U1 and U2?

    • Non-uniform skin distribution. Stab 1: "false" stabilization due to the non-uniform skin distribution or radial flow in the low skin layer. Stab 2: radial flow regime for the total system and Stab 3: "false" stabilization due to distortion. 2*K1H1 = 5506.8 mD.ft, hence K1= 50.1mD and K2= 76.5mD
    • Non-uniform skin distribution. Stab 1: "false" stabilization due to the non-uniform skin distribution or radial flow in the low skin layer. Stab 2: reservoir crossflow and Stab 3: radial flow regime for the total system. We have: 2*K1H1 = 3671.2 mD.ft, hence K1= 33.4mD and K2= 51.0mD
    • Uniform skin distribution. Stab 1: radial flow for the total system, Stab 2: reservoir heterogeneity (increase in mobility) and Stab 3: "false" stabilization due to distortion. 2*K1H1 = 2753.4 mD.ft, hence: K1= 25.0mD and K2= 38.2mD
    • Uniform skin distribution. Stab 1: radial flow in high permeability layer, Stab 2: reservoir crossflow and Stab 3: radial flow for the total system. 2*K1H1 = 3671.2 mD.ft, hence: K1= 33.4mD and K2= 51.0mD
    Correct
    

    Non-uniform skin distribution (in this case, high skin in one layer).
    1: false stabilization due to the non-uniform skin distribution or radial flow in the low skin layer
    2: radial flow regime for the total system
    3: false stabilization due to distortion.
    2 K1H1 = 5506.8 mD.ft -> K1= 50.1mD and K2= 76.5mD

    Incorrect
    
14. Question 14 of 15

    14. Question

     

    From the distributed temperature array, we might conclude that 90% of the production comes from the top layer U1. If we assume an uniform skin distribution across the perforated intervals, what could the derivative stabilizations 1, 2 and 3 represent? Could we extract the KH and permeability values in layers U1 and U2 in this case?

    • Stab 1: radial flow in the total system, Stab 2: reservoir crossflow from layer U2 to U1 or increase in mobility and Stab 3: false stabilization due to distortion. KH= 2753.4 mD.ft, hence: K1H1= 45.1mD and K2= 7.6mD
    • Stab 1: radial flow in the high permeability layer, Stab 2: reservoir crossflow from layer U2 to U1, and Stab 3: radial flow for the total system. We have total KH = 3671.2 mD.ft and K1H1= 0.9*KH= 3304.1 mD.ft and K2H2= 367.1 mD.ft. Hence K1= 60.1mD and K2= 10.2mD.
    • Stab 1: radial flow in the high permeability layer, Stab 2: increase in mobility and Stab 3: false stabilization due to distortion. We have K1H1 = 2753.4 mD.ft and therefore K2H2= 0.1* K1H1/0.9 = 305.9mD.ft. Hence K1= 50.1 and K2= 8.5mD.
    • Stab 1: radial flow in the low permeability layer, Stab 2: crossflow in the reservoir from layer U2 to U1, and Stab 3: radial flow for the total system. We have: K2H2 = 2753.4 mD.ft and therefore K1H1= 0.9 * K2H2/0.1= 24780.6 mD.ft. Hence K1= 450.6 mD and K2= 76.5mD.
    Correct
    

    Incorrect
    
15. Question 15 of 15

    15. Question

     

    In hindsight, how could we have reduced the flaring and shut-in duration? How many hours could we have approximately saved? (multiple choice)

    (Hint: we can extract the minimum connected oil volume at surface with the following equation:
    
    with Vmin: the minimum connected oil volume at surface in barrels, Sw: the water saturation, q: the oil rate in STB/D, Ct: the total compressibility in 1/psi, ∆t and ∆p’ the elapsed time in hours and derivative value in psi at the start of the pseudo-steady state).

    • We could have reduced flaring with a better test design and used a real-time deconvolution during the operations for the shut-in duration.
    • We could have reduced flaring using a DP of a few psi and optimized PBU 2 using the real-time conventional derivative.
    • We could have saved 78 hours or about 3 days.
    • We could have removed PBU2 test, as we had an ESD and PBU 1. As a result, we could have saved about 90 hours, or 3.8 days
    • We could have saved about 24 hours from PBU 2 test.
    Correct
    

    We could have reduced flaring with a better test design (using the trial modern techniques) and ensuring that the pressure change is larger than the gauge resolution/noise. We could have used a real-time deconvolution for the shut-in duration.
    qdt/dp’ = 2837.6 stb.hr/psi -> dt ~50hrs. We could have saved 128-50= 78 hours or about 3 days.

    Incorrect
    

YC1 exploration well test with two DST performed in a commingled oil reservoir

Well Testing in an exploration well with the use of Deconvolution to calculate the minimum connected oil volume