quarterly research performance progress report (period ... · the eleventh quarterly report was...
TRANSCRIPT
1
DOE Award No.: DE-FE0028973
Quarterly Research Performance Progress
Report
(Period Ending 9/30/2019)
Advanced Simulation and Experiments of
Strongly Coupled Geomechanics and Flow for
Gas Hydrate Deposits: Validation and Field
Application Project Period (10/01/2016 to 12/31/2019)
Submitted by:
Jihoon Kim
The Harold Vance Department of Petroleum Engineering,
College of Engineering
Texas A&M University
501L Richardson Building
3116 College Station TX, 77843-3136
Email: [email protected]
Phone number: (979) 845-2205
Prepared for:
United States Department of Energy
National Energy Technology Laboratory
OIL & GAS
September 30, 2019
Office of Fossil Energy
2
DISCLAIMER
This report was prepared as an account of work sponsored by an agency of the United States
Government. Neither the United States Government nor any agency thereof, nor any of their
employees, makes any warranty, express or implied, or assumes any legal liability or
responsibility for the accuracy, completeness, or usefulness of any information, apparatus,
product, or process disclosed, or represents that its use would not infringe privately owned
rights. Reference herein to any specific commercial product, process, or service by trade name,
trademark, manufacturer, or otherwise does not necessarily constitute or imply its endorsement,
recommendation, or favoring by the United States Government or any agency thereof. The views
and opinions of authors expressed herein do not necessarily state or reflect those of the United
States Government or any agency thereof.
3
TABLE OF CONTENTS
Page
DISCLAIMER ................................................................................................................. 2
TABLE OF CONTENTS ................................................................................................. 3
ACCOMPLISHMENTS ................................................................................................... 4
Objectives of the project ............................................................................................ 4
Accomplished ............................................................................................................ 4
Task 1 ....................................................................................................................... 4
Task 2 ....................................................................................................................... 5
Task 3 ....................................................................................................................... 5
Task 4 ..................................................................................................................... 11
Task 5 ..................................................................................................................... 13
Task 6 ..................................................................................................................... 16
PRODUCTS ................................................................................................................. 17
BUDGETARY INFORMATION ..................................................................................... 17
4
ACCOMPLISHMENTS
Objectives of the project
The objectives of the proposed research are (1) to investigate geomechanical responses induced
by depressurization experimentally and numerically; (2) to enhance the current numerical
simulation technology in order to simulate complex physically coupled processes by
depressurization and (3) to perform in-depth numerical analyses of two selected potential
production test sites: one based on the deposits observed at the Ulleung basin UBGH2-6 site; and
the other based on well-characterized accumulations from the westend Prudhoe Bay. To these
ends, the recipient will have the following specific objectives:
1). Information obtained from multi-scale experiments previously conducted at the recipient’s
research partner (the Korean Institute of Geoscience and Mineral Resources (KIGAM)) that were
designed to represent the most promising known Ulleung Basin gas hydrate deposit as drilled at
site UBGH2-6 will be evaluated (Task 2). These findings will be further tested by new
experimental studies at Lawrence Berkeley National Laboratory (LBNL) and Texas A&M (TAMU)
(Task 3) that are designed capture complex coupled physical processes between flow and
geomechanics, such as sand production, capillarity, and formation of secondary hydrates. The
findings of Tasks 2 and 3 will be used to further improve numerical codes.
2) Develop (in Tasks 4 through 6) an advanced coupled geomechanics and non-isothermal flow
simulator (T+MAM) to account for large deformation and strong capillarity. This new code will be
validated using data from the literature, from previous work by the project team, and with the
results of the proposed experimental studies. The developed simulator will be applied to both
Ulleung Basin and Prudhoe Bay sites, effectively addressing complex geomechanical and
petrophysical changes induced by depressurization (e.g., frost-heave, strong capillarity, cryo-
suction, induced fracturing, and dynamic permeability).
Accomplished
The plan of the project timeline and tasks is shown in Table 1, and the activities and achievements
during this period are listed in Table 2.
Task 1: Project management and planning
The eleventh quarterly report was submitted to NETL on July 29, 2019. LBNL and TAMU continued
to work on Subtasks 3.3 and 3.4, respectively. For Subtask 3.4, a TAMU student visited LBNL for
the experiment on September. Meanwhile, the TAMU-KIGAM team have mainly been working
on Subtasks 4.3 and 5.6 as well as Task 6. The specific status of the milestones is shown in Table
2. We had requested no-cost extension of the project to December 31 2019 and it was approved
August 22 2019. Specific achievements including presentation and publication during this period
are as follows.
5
Task 2: Review and evaluation of experimental data of gas hydrate at various scales for gas
production of Ulleung Basin
Subtask 2.1 Evaluation of Gas hydrate depressurization experiment of 1-m scale
This task was completed.
Subtask 2.2 Evaluation of Gas hydrate depressurization experiment of 10-m scale
This task was completed.
Subtask 2.3 Evaluation of Gas hydrate depressurization experiment of 1.5-m scale system in 3D
This task was completed.
Subtask 2.4 Evaluation of gas hydrate production experiment of the centimeter-scale system
This task was completed.
Task 3: Laboratory Experiments for Numerical Model Verification
Subtask 3.1: Geomechanical changes from effective stress changes during dissociation
This task was completed.
Subtask 3.2 Geomechanical changes from effective stress changes during dissociation – sand
This task was completed.
Subtask 3.3 Geomechanical changes resulting from secondary hydrate and capillary pressure
changes
During this quarter, methane gas hydrate was formed using the excess gas method in a sample
consisting of F110 sand packed in an elastomer sleeve located inside of a LBNL custom hydrate
core holder. The apparatus was equipped with multiple ports to accommodate fluid flow,
pressure and temperature monitoring, and three capillary pressure sensors. The capillary
pressure sensors were constructed from porous ceramic with length 0.5 in, diameter 0.25 in, and
had a nominal 5 bar (72.5 psi) air entry pressure (Soil moisture, Santa Barbara, CA). A 1/16 in
hole was drilled into one end and a nylon tube was epoxied into the hole. Before installing in the
6
vessel, the sensors and tubing were saturated with water. The sensors were packed with the
sand and were located near the sample inlet, mid sample, and near the outlet. Thermocouples
were located in the confining fluid, with the exception of one located in the inlet line, and one
located in the outlet line both in close proximity to the sample. The sample is 5.7 in (14.5 cm)
long and 2 in (5 cm) in diameter. A 3/16 in aluminum tube coil connected to a circulating
temperature controller was placed around the sample at the outlet end to enable establishing a
temperature gradient in (See Fig. 3.3.1). Note that two of the capillary sensors were located in
the region with the warming coil and the third sensor was located outside the coil near the inlet.
Fig 3.3.1 Top – entire system showing sample, aluminum coil, and inlet and outlet plumbing. Red
dots indicate locations of thermocouples. Bottom – cut out of sample only, showing location of
capillary pressure sensors.
After the moist sand was packed in the sleeve, confining and pore pressure were set to 800 psi
and 700 psi, respectively, and the temperature was chilled to the hydrate stability region, 4 °C,
to allow hydrate to form. After hydrate formation the sample was saturated with water and a
temperature gradient was imposed on the sample such that one end of the sample was outside
of hydrate equilibrium, while the other end is still in equilibrium. Because the majority of
thermocouples were located in the confining fluid, not the sample, these are an approximation
of the sample temperature (Fig. 3.3.2). During the temperature modifications, effects of the
secondary hydrate formation resulting from the thermal dissociation was monitored, and the
capillary pressures generated in the process. The first temperature gradient was applied on day
outlet
Aluminum Coil
inlet
7
6, when the temperature of the fluid in the aluminum coil was raised to a region in the sample
that was outside the hydrate stability zone for approximately 1 day, then lowered again so the
entire system was in the stable hydrate zone. This was repeated 3 more times, for a total of 4
cycles over a period of 30 days (Fig. 3.3.3). Before final dissociation, the system was set at a ‘mid’
point temperature.
Fig. 3.3.2 Temperature profiles along sample. Position of the thermocouples are measured
relative to the endcap next to the sample.
8
Fig. 3.3.3 Temperature profiles in sample. Positions of the thermocouples are measured from
the inlet endcap of the large vessel. Dashed line indicates temperature where hydrate becomes
unstable (pore pressure 700 psi)
The capillary pressure sensors were connected to differential pressure transducers (DPT), and
the sample inlet. Fig. 3.3.4 shows capillary pressures at the sensors as measured by differential
pressure transducers. Without hydrate and during warming, the DPTs showed low values
indicating that water was freed from the hydrate. When the sample was cold and hydrate was
present, the DPTs showed high capillary pressure indicating the sample was “dry” (cryosuction)
due to water forming hydrate. During the mid-temperature setting, a different pattern was
observed in the sensors, which was likely due to a pressure drop in the inlet, and may have been
unrelated to hydrate formation. The three sensors behaved similarly. The sensor DelP mid was
connected to a sensor with a maximum 10 psi differential which resulted in lower apparent
readings for that position. The magnitudes of the capillary pressures were not expected to exceed
10 psi, however they greatly exceeded that value.
Fig. 3.3.4 Capillary pressures at the sensors as measured by differential pressure transducers.
During the warming and cooling processes, X-ray CT was used to monitor methane hydration
formation in the sample (Fig. 3.3.5). The darker purple is lower density and brighter yellow is high
-20
0
20
40
60
80
0 5 10 15 20 25 30 35
del
ta P
(p
si)
time (days)
Capillary sensors
DelP inlet DelP mid Del P outlet
warming
cooling
9
density. As hydrate forms, the density of the sample increases. Comparing panels Figs. 3.3.5 (b)
and (c), hydrate dissociation during warming is shown by the lower density region to the right.
Fig. 3.3.6 shows average z-axis profiles during warm and cold states, or the average density of
each slice along the sample during the experiment, which also shows the density changes
occurring along the sample.
Despite differences in hydrate conditions in the sample, the capillary sensors did not show
differences in the values depending on position within the sample. One exception is during the
mid-temperature cycle during the end of the experiment, there seemed to be some variation in
the values of the three sensors. This was probably due to some pressure loss in the inlet pump,
which caused some pressure differences across the sample at around day 32.
Operationally this test was complicated to set up taking more than a week, therefore once built
having the opportunity to perform multiple measurements on one build is advantageous. Future
tests could include potentials at more temperature gradients, different initial water saturations,
or layers of samples with different grain sizes. In addition, it would be helpful to have more
temperature measurements within the sample to more accurately define the temperature profile
to be able to control and understand the proximity to the equilibrium point. In addition, water
potentials in the sample were higher than expected. New tests with this system should include
differential pressure transducers that can measure ranges up to 80 psi.
A number of future analyses should be performed both on these data, and using this new
capability. The declining capillary pressure over time is curious. Does this occur with hydrate
ripening; is it an experimental artifact; are there other reasons? Closer investigations of the data,
particularly during transitions will provide a better description of transient system behavior
showing the effect of temperature gradient and simultaneously conditions across the equilibrium
point.
(a) Hydrate formed in sample, before water saturation
10
(b) Sample during cold cycle
(c) Sample during warming cycle
(d) Sample at mid-temperature
Fig. 3.3.5 X-ray cross sections of sample under different conditions (a) hydrate formation (b)
water saturated, cold (c) water saturated, warm (d) mid-temperature.
11
Fig. 3.3.6. Z-axis profiles of CT intensity of the sample during warm and cool periods. For clarity,
z-axis profiles were averaged from the repeated cycles. The warm average corresponds to
condition in Fig. 3.3.4 warming cycles and Fig. 3.3.5c. Similarly, cold average corresponds to
cooling cycles in Fig. 3.3.4 and to the CT scan in Fig. 3.3.5b. The mid-point temperature is from a
single sample, corresponding to Fig. 3.3.5d. Intensity is in Hounsfield Units.
Subtask 3.4 Construction of the Relative Permeability Data in Presence of Hydrate
While we are working on this task, we have not obtained visible outcomes, yet.
Subtask 3.5 Identification of Hysteresis in Hydrate Stability
This subtask was competed.
Task 4: Incorporation of Laboratory Data into Numerical Simulation Model
Subtask 4.1 Inputs and Preliminary Scoping Calculations
This task is completed. We have finished revisiting and reviewing the experimental data of
Subtasks 2.1 and 2.3 as well as the input data for Subtasks 5.5 and 5.6 and Task 6. In particular,
we have reflected the input data of the Ulleung Basin model updated by KIGAM in 2017. Thus,
we have the two models developed in 2014 and 2017, respectively, which are being used for
simulation.
12
Subtask 4.2 Determination of New Constitutive Relationships
No further progress has been made during the quarter. New constitutive relationships can be
obtained after Subtask 3.4.
Subtask 4.3 Development of Geological Model
Continuing to the previous research, along with Subtask 4.1, we have updated the geological
model of UBGH2-6 for simulation the flow and geomechanics problems (Fig. 4.3.1). During this
quarter, we have focused on constructing the new input and mesh files for T+H simulation to
simulate the UBGH2-6 2017 model.
We took the UBGH2-6 2014 model for Subtask 5.6 in order to analyze productivity of gas and
subsidence by using TOUGH+ROCMECH, fast and efficient, when considering various production
scenarios.
At the same time, we are taking the UBGH2-6 2017 model for Subtask 5.6 in order to investigate
shear slip and stress concentration near the casing and bottom of the well, respectively, by using
TOUGH+FLAC3D, which allows more complex gridding in 3D.
Currently, we are making the geological model of PBU-L-106C (Kim et al. 2012, Journal of
Petroleum Science and Engineering, 92-93: 143-157) for the same analyses as we did for the
UBGH2-6 case by using both TOUGH+ROCMECH and TOUGH+FLAC3D.
13
Fig. 4.3.1. A new UBGH2-6 model updated in 2017. There are thin hydrate layers that have low
gas hydrate saturation, which yields high effective permeability due to low solid phase saturation.
Task 5: Modeling of coupled flow and geomechanics in gas hydrate deposits
Subtask 5.1 Development of a coupled flow and geomechanics simulator for large deformation
This task was completed.
Subtask 5.2 Validation with experimental tests of depressurization
In the previous progress reports, we showed good agreement between the lad data and
simulation results, which validates TOUGH+ROCMECH. We hence mark this subtask completed.
We plan to further calibrate the simulation models by tuning initial condition and other flow and
geomechanics properties.
Subtask 5.3 Modeling of sand production and plastic behavior
We have been implementing to TOUGH+ROCMECH sand production models that are based on
elastoplastic geomechanics. We will complete this subtask next quarter.
Subtask 5.4 Modeling of induced changes by formation of secondary hydrates: Frost-heave,
strong capillarity, and induced fracturing
We have been implementing to TOUGH+ROCMECH a new model for hysteretic capillary pressure
and relative permeability. This modeling approach is on the basis of strong mathematical analysis
used in the return mapping algorithms of elastoplasticity, being thermodynamically consistent.
The details are described in the paper recently accepted in Journal of Computational Physics
during this quarter, shown in the PRODUCT section. We are currently implementing the van
Genuchten model as follows.
Van Genuchten model:
𝑆�̅� = [1 + (𝛼𝑃𝑐)𝑛]−𝑚 ⇔ 𝑃𝑐 =1
𝛼[𝑆�̅�
−1
𝑚 − 1]1
𝑛, 𝑘𝑟𝑔 = 𝑆�̅�1/2
[1 − (1 − 𝑆�̅�
1
𝑚)𝑚]2
,
where 𝑆�̅� =𝑆𝐽−𝑆𝑟𝐽
1−𝑆𝑟𝐽 is the effective saturation of J phase and 𝛼, m, n, 𝜆 are the model
parameters, where m=1-1/n.
14
Fig. 5.4.1. Hysteresis modeling for the van Genuchten model. Left: Capillary pressure. Right:
Relative permeability of gas phase. n=2.0, and Srw=0. α=0.0001 for the reversible (elastic) process,
while α=0.0002 for the irreversible (plastic) process.
We will complete this subtask next quarter after additionally modifying hydraulic fracturing
modeling modules to TOUGH+ROCMECH.
Subtasks 5.5 and 5.6 Field-scale simulation of PBU L106 and Ulleung Basin
We studied productivity of gas as well as geomechanical responses subsidence for 4 cases shown
in Table 5.5.1 by using TOUGH+ROCMECH when considering 30 day production. The simulation
is based on the geological model developed in 2014. Here, the brief summary of the simulation
results is as follows.
Figs. 5.5.1 and 5.5.2 show production of gas and water for the four cases. Overall, for all cases,
gas production is much higher than water production. When BHP (bottom hole pressure) is low
(i.e., Case 1), we can get higher production of gas and water while larger subsidence is obtained.
For Case 2, where BHP drops linearly rather than instantaneously, the numerical results of
production are still almost same as those of the reference case, where BHP is 9MPa. For Case 3,
where periodic BHP’s of 9MPa and 14MPa are applied, we can have almost same results as those
of reference case. On the other hand, from Fig. 5.5.3, the subsidence of Case 3 is the lowest
among the four cases, which shows that this case is one of the promising scenarios for UBGH2-6
field test production.
Table 5.5.1. Four production scenarios for 30 day production by depressurization with a vertical
well.
15
Fig. 5.5.1. Left: methane gas flow rate. Right: cumulative production of gas phase methane.
Table 5.5.2 Total gas production for the four cases.
Fig. 5.5.2. Left: water flow rate. Right: cumulative production of water.
Table 5.5.3 Total water production for the four cases.
16
Fig. 5.5.3. Vertical displacements near and away from the well for all four cases. Overall,
subsidence is not significant during 30 day production.
Task 6: Simulation-Based Analysis of System Behavior at the Ignik-Sikumi and Ulleung Hydrate
Deposits
We are performing numerical simulation on the wellbore slip and stress concentration near the
well for UBGH2-6. The simulation is based on the geological model developed in 2017. Unlike the
2014 model, we identify that there is fast gas hydrate dissociation at the layers that have low gas
hydrate saturation due to high permeability. Also, we find stress concentration at the bottom of
the well. We are currently analyzing these behavior more, considering different production
scenarios.
17
PRODUCTS
Publication
Journal paper
Yoon H.C., Zhou P. , Kim J., 2019, Robust modeling of hysteretic capillary pressure and relative
permeability for two phase flow in porous media, Journal of Computational Physics, Accepted
The fund was acknowledged in this paper.
BUDGETARY INFORMATION
Table 3 shows the information of the budget for this project and the expenditure up to
9/30/2019. The expenditure by TAMU and cost-share from KIGAM are accurate while the
expenditure by LBNL might not be accurate. For detailed information of the budget and
expenditure, refer to the financial status report separately submitted to NETL by each institution.
Table 1 – Initial project timeline and milestones (Gantt Chart)
FY17 FY18 FY19
Quarter Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Task 1.0. Project Management/Planning A
Task 2.0. Experimental study of gas hydrate in
various scales for gas production of Ulleung
Basin
Subtask 2.1. Depressurization of 1 m scale in 1D B Subtask 2.2 Depressurization of 10-m scale in 1D C Subtask 2.3. Depressurization of 1.5-m scale in 3D D Subtask 2.4. Revisit to the centimeter-scale system
Task 3.0. Laboratory Experiments for
Numerical Model Verification
Subtask 3.1. Effective stress changes during dissociation E Subtask 3.2. Sand production F Subtask 33. Secondary hydrate and capillary pressure changes
G
Subtask 3.4. Relative Permeability Data Subtask 3.5. Hysteresis in Hydrate Stability
Task 4.0. Incorporation of Laboratory Data
into Numerical Simulation Model
Subtask 4.1. Inputs and Preliminary Scoping Calculations H Subtask 4.2. Determination of New Constitutive Relationships
18
Subtask 4.3. Development of Geological Model
Task 5.0. Modeling of coupled flow and
geomechanics in gas hydrate deposits
Subtask 5.1 Development of a coupled flow and geomechanics
simulator for large deformation I
Subtask 5.2 Validation with experimental tests of depressurization
J
Subtask 5.3 Modeling of sand production and plastic behavior K Subtask 5.4 Frost-heave, strong capillarity, and induced
fracturing L
Subtask 5.5 Field-scale simulation of PBU L106 Subtask 5.6 Field-wide simulation of Ulleung Basin
Task 6.0. Simulation-Based Analysis of System
Behavior at the Ignik-Sikumi and Ulleung
Hydrate Deposits
M
Table 2. Milestones Status
Milestone Description Planned
Completion
Actual
Completion
Status / Comments
Task 1 Milestones
Milestone A Complete the kick-off meeting
and revise the PMP
12/31/17 1/14/2017 Kickoff meeting held
11/22/17, revised PMP
finalized 1/17/17
Task 2 Milestones
Milestone B Complete analysis of 1 m-
scale experiment in 1D and
validation of the cm-scale
system (FY17, Q4)
9/30/2017 Completed.
Milestone C Complete analysis of 10m-
scale experiment in 1D
6/30/2018 Completed.
Milestone D Complete analysis of 1.5m-
scale experiment in 3D
Completed.
Task 3 Milestones
Milestone E Complete geomechanical
changes from effective stress
changes during dissociation
and construction of the
relative permeability data
9/30/2017 Completed
Milestone F Complete geomechanical
changes from effective stress
changes during dissociation
(sand production) and
hysteresis in hydrate stability
9/30/2018 Completed
Milestone G Complete geomechanical
changes resulting from
secondary hydrate and
capillary pressure changes
9/30/2019 Ongoing
Task 4 Milestones
Milestone H Complete inputs and
preliminary scoping
calculations, determination of
New Constitutive
12/31/2018 Ongoing
19
Relationships, development of
Geological Model
Task 5 Milestones
Milestone I Complete development of a
coupled flow and
geomechanics simulator for
large deformation, validation
with experimental tests of
Subtasks 2.1 and 2.4.
9/30/17 Completed
Milestone J Validation with experimental
tests of Task 2 and 3
3/31/2019 Completed
Milestone K Complete modeling of sand
production and plastic
behavior, validation with
experimental tests of Subtasks
3.3
9/30/2018 Ongoing
Milestone L Complete field-scale
simulation of the Ulleung
Basin and PBU L106
9/30/2019 Ongoing
Task 6 Milestones
Milestone M Complete Task 6 9/30/2019 Ongoing
Table 3 Budget information
20
Q1 Cumulative Total Q2 Cumulative Total Q3 Cumulative Total Q4 Cumulative Total
Baselinie Cost Plan
Federal (TAMU) $43,543 $43,543 $36,189 $79,733 $47,526 $127,259 $41,209 $168,468
Federal (LBNL) $18,750 $18,750 $18,750 $37,500 $18,750 $56,250 $18,750 $75,000
Non-Federal Cost Share $6,986 $6,986 $6,986 $13,972 $6,986 $20,958 $6,986 $27,944
Total Planned $69,279 $69,279 $61,925 $131,205 $73,262 $204,467 $66,945 $271,412
Actual Incurred Cost
Federal (TAMU) $46,338 $46,338 $47,068 $93,406 $32,930 $126,336 $48,234 $174,570
Federal (LBNL) $6,658 $6,658 $39,707 $46,365 $16,775 $63,140 $67,711 $130,851
Non-Federal Cost Share $6,986 $6,986 $6,986 $13,972 $6,986 $20,958 $6,986 $27,944
Total incuured cost $59,982 $59,982 $93,761 $153,743 $56,691 $210,434 $122,931 $333,365
Variance
Federal (TAMU) $2,795 $2,795 $10,878 $13,673 ($14,596) ($923) $7,025 $6,102
Federal (LBNL) ($12,092) ($12,092) $20,957 $8,865 ($1,975) $6,890 $48,961 $55,851
Non-Federal Cost Share $0 $0 $0 $0 $0 $0 $0 $0
Total variance ($9,297) ($9,297) $31,835 $22,538 ($16,571) $5,967 $55,986 $61,953
Baselinie Reporting Quarter
Budget Period 3
Q1 Q2 Q3 Q4
10/01/18-12/31/18 01/01/19-03/31/19 04/01/19-06/30/19 07/01/19-09/30/19
National Energy Technology Laboratory 626 Cochrans Mill Road P.O. Box 10940 Pittsburgh, PA 15236-0940 3610 Collins Ferry Road P.O. Box 880 Morgantown, WV 26507-0880 13131 Dairy Ashford Road, Suite 225 Sugar Land, TX 77478 1450 Queen Avenue SW Albany, OR 97321-2198 Arctic Energy Office 420 L Street, Suite 305 Anchorage, AK 99501 Visit the NETL website at: www.netl.doe.gov Customer Service Line: 1-800-553-7681