Quest Pipelines Flow and Flow Assurance Design …Quest Pipelines Flow and Flow Assurance Design and...

Disclaimer

This Report, including the data and information contained in this Report, is provided to you on an

“as is” and “as available” basis at the sole discretion of the Government of Alberta and subject to the

terms and conditions of use below (the “Terms and Conditions”). The Government of Alberta has

not verified this Report for accuracy and does not warrant the accuracy of, or make any other

warranties or representations regarding, this Report. Furthermore, updates to this Report may not

be made available. Your use of any of this Report is at your sole and absolute risk.

This Report is provided to the Government of Alberta, and the Government of Alberta has obtained

a license or other authorization for use of the Reports, from:

Shell Canada Energy, Chevron Canada Limited. and Marathon Oil Canada Corporation, for

the Quest Project

(collectively the “Project”)

Each member of the Project expressly disclaims any representation or warranty, express or

implied, as to the accuracy or completeness of the material and information contained herein, and

none of them shall have any liability, regardless of any negligence or fault, for any statements

contained in, or for any omissions from, this Report. Under no circumstances shall the Government

of Alberta or the Project be liable for any damages, claims, causes of action, losses, legal fees or

expenses, or any other cost whatsoever arising out of the use of this Report or any part thereof or

the use of any other data or information on this website.

Terms and Conditions of Use

Except as indicated in these Terms and Conditions, this Report and any part thereof shall not be

copied, reproduced, distributed, republished, downloaded, displayed, posted or transmitted in any

form or by any means, without the prior written consent of the Government of Alberta and the

Project.

The Government of Alberta’s intent in posting this Report is to make them available to the public

for personal and non-commercial (educational) use. You may not use this Report for any other

purpose. You may reproduce data and information in this Report subject to the following

conditions:

• any disclaimers that appear in this Report shall be retained in their original form and

applied to the data and information reproduced from this Report

• the data and information shall not be modified from its original form

• the Project shall be identified as the original source of the data and information, while this

website shall be identified as the reference source, and

• the reproduction shall not be represented as an official version of the materials reproduced,

nor as having been made in affiliation with or with the endorsement of the Government of

Alberta or the Project

By accessing and using this Report, you agree to indemnify and hold the Government of Alberta and

the Project, and their respective employees and agents, harmless from and against any and all

claims, demands, actions and costs (including legal costs on a solicitor-client basis) arising out of

any breach by you of these Terms and Conditions or otherwise arising out of your use or

reproduction of the data and information in this Report.

Your access to and use of this Report is subject exclusively to these Terms and Conditions and any

terms and conditions contained within the Report itself, all of which you shall comply with. You will

not use this Report for any purpose that is unlawful or prohibited by these Terms and Conditions.

You agree that any other use of this Report means you agree to be bound by these Terms and

Conditions. These Terms and Conditions are subject to modification, and you agree to review them

periodically for changes. If you do not accept these Terms and Conditions you agree to immediately

stop accessing this Report and destroy all copies in your possession or control.

These Terms and Conditions may change at any time, and your continued use and reproduction of

this Report following any changes shall be deemed to be your acceptance of such change.

If any of these Terms and Conditions should be determined to be invalid, illegal or unenforceable

for any reason by any court of competent jurisdiction then the applicable provision shall be severed

and the remaining provisions of these Terms and Conditions shall survive and remain in full force

and effect and continue to be binding and enforceable.

These Terms and Conditions shall: (i) be governed by and construed in accordance with the laws of

the province of Alberta and you hereby submit to the exclusive jurisdiction of the Alberta courts,

and (ii) ensure to the benefit of, and be binding upon, the Government of Alberta and your

respective successors and assigns.

Project

Document Title

Document Number

Document Revision

Document Status

Document Type

Control ID

Owner / Author

Issue Date

Expiry Date

ECCN

Security Classification

Disclosure

Revision History shown on next page

Heavy Oil

Controlled Document

Quest CCS Project

Quest Pipelines Flow and Flow Assurance Design and Operability Report

Quest CCS Project

Quest Pipelines Flow and Flow Assurance Design and

Operability Report

07-2-LA-5507-0003

Rev 01

Approved

LA5507-Design Philosophy

248

Leonid Dykhno

2011-08-17

None

EAR 99

None

07-2-LA-5507-0003

Revision History

REVISION STATUS APPROVAL

Rev. Date Description Originator Reviewer Approver

01

2011-08-15

Issued for Review

Leonid Dykhno

All signed originals will be retained by the UA Document Control Center and an electronic copy will be stored in Livelink

Signatures for this revision

Date

Role

Name Signature or electronic reference (email)

Originator Leonid Dykhno

Reviewer

Approver

Summary

Keywords

Quest Pipelines Flow and Flow Assurance Design

and Operability Report

Rev 01

Heavy Oil

07-2-LA-5507-0003

TABLE OF CONTENTS

SUMMARY ................................................................................................................................ 7

1.0 BACKGROUND ............................................................................................................. 8

2.0 FLOW ASSURANCE STRATEGIES.............................................................................10

2.1 Solids Management ...................................................................................................10

2.1.1 Hydrates ................................................................................................................10

2.1.2 Wax .......................................................................................................................10

2.1.3 Pour point ............................................................................................... ...............10

2.1.4 Asphaltenes ...........................................................................................................10

2.1.5 Scale .....................................................................................................................10

2.1.6 Corrosion ...............................................................................................................10

2.1.7 Emulsions ............................................................................................... ...............11

2.1.8 Slugging ................................................................................................................11

2.1.9 Injected Solids ............................................................................................... ........11

2.1.10 Chilly Choke............................................................................................... ........11

2.2 Operational Considerations .......................................................................................11

2.2.1 Start-Up .................................................................................................................11

2.2.2 Steady-state ............................................................................................... ...........12

2.2.3 Shut-In ...................................................................................................................12

2.2.4 Flowline Venting ....................................................................................................12

3.0 BASIC DATA ................................................................................................................13

3.1 PVT and Reservoir Data ............................................................................................13

3.2 Fluid Compositions ....................................................................................................14

3.3 Water Samples ..........................................................................................................15

3.4 Hydrates ....................................................................................................................15

3.4.1 Hydrate Inhibition Requirements ............................................................................16

3.5 Chilly Choke ..............................................................................................................18

3.6 Wax and Pour Point...................................................................................................20

3.7 Asphaltenes...............................................................................................................20

3.8 Scale .........................................................................................................................20

3.9 Well Details ...............................................................................................................20

3.10 Flowline Details .........................................................................................................21



Rev 01

Heavy Oil

07-2-LA-5507-0003

3.11 Production Function...................................................................................................23

4.0 STEADY STATE ANALYSIS ........................................................................................24

4.1 Low Flow Events ............................................................................................... ........24

4.1.1 Wellbore Flowing Conditions..................................................................................24

4.1.2 Hydrate Risk ............................................................................................... ...........27

4.1.3 Hydrate Mitigation ..................................................................................................29

4.1.4 Flow Assurance Risks in Pipeline ..........................................................................30

4.1.5 Conclusions ...........................................................................................................30

4.2 Normal Production – Temperatures for Above-Ground Piping Sections.....................31

4.2.1 Figures for Steady State Temperatures at all Line Break Valves ...........................32

4.2.2 Figures for well pad locations 1 and 5 ....................................................................38

4.3 Normal Production – Flowline and Well Operating Envelopes ...................................40

5.0 TRANSIENT ANALYSIS ...............................................................................................47

5.1 Fluid Hammer ............................................................................................... .............47

5.1.1 Fluid Hammer in the Wellbore................................................................................48

5.1.2 Fluid Hammer in Flowline ......................................................................................57

5.2 Flowline Venting ............................................................................................... .........61

5.2.1 Description of heat transfer in model......................................................................61

5.2.2 Simplified flowline model for screening ..................................................................64

5.2.3 Detailed flowline model for venting ........................................................................69

5.2.4 Venting of Flowline Sections between LBVs ..........................................................75

5.3 Wellbore Venting and Placement of Subsurface Safety Valve ...................................77

5.3.1 Blowout in a well with a low reservoir productivity ..................................................81

6.0 REFERENCES ..............................................................................................................83

List of Figures

Figure 3.1 Geothermal temperature profile for wells (4) ............................................................14

Figure 3.2 Model prediction and comparison with experimental data (5) ...................................16

Figure 3.3 Predicted hydrate curve for composition during normal operation ............................16

Figure 3.4 Methanol requirement for hydrate inhibition – Impact of temperature .......................17

Figure 3.5 Methanol requirement for hydrate inhibition – Impact of water content .....................18

Figure 3.6 Hydrate formation in relation to JT cooling across well choke (MultiFlash) ...............19

Figure 3.7 Hydrate formation in relation to JT cooling across well choke (STFlash) ..................20

Figure 3.8 Detailed flowline topography with location of LBVs and well branches .....................22

Figure 4.1 Flowing bottomhole pressure as a function of CO2 injection rate ..............................25



Rev 01

Heavy Oil

07-2-LA-5507-0003

Figure 4.2 Wellhead pressure (downstream of choke) as a function of the CO2 rate .................26

Figure 4.3 CO2 density as a function of temperature .................................................................26

Figure 4.4 Wellbore temperature at different CO2 injection rates with hydrate curve .................27

Figure 4.5 Hydrate risk at 4 lbs H2O/MMscf water content ........................................................28

Figure 4.6 Hydrate risk at 6 lbs H2O/MMscf water content ........................................................28

Figure 4.7 Temperature at LBV-1 during steady state operation ...............................................33



Figure 4.10 Temperature at LBV-4 during steady state operation .............................................35




Figure 4.14 Impact of heat transfer coefficient on the temperature at LBV-1 .............................38

Figure 4.15 Arrival temperature at Well 1 ..................................................................................39

Figure 4.16 Arrival temperature at Well 5 ..................................................................................40

Figure 4.17 Schematic for operating lines upstream and downstream of well choke .................41

Figure 4.18 Operating line for Well 1 with the normal composition ............................................42

Figure 4.19 Operating line for Well 1 with the upset composition ..............................................43

Figure 4.20 Operating line for Well 5 with the normal composition ............................................44

Figure 4.21 Operating line for Well 5 with the upset composition ..............................................44

Figure 5.2 Maximum surge pressure in well upon shut-in..........................................................50

Figure 5.3 Minimum surge pressure in well upon shut-in...........................................................51

Figure 5.4 Increase in pressure above steady state values (Base Case) ..................................52

Figure 5.5 Increase in pressure above steady state values (low PI case)..................................53

Figure 5.6 Increase in pressure above steady state values (Low Injection Rate).......................54

Figure 5.7 Increase in pressure above steady state values (High Temperature Case-Summer)55

Figure 5.7 Change in pressure above and below SC-SSSV during blowout scenario...............56

Figure 5.8 Pressure surges upon closing LBV-1 .......................................................................58

Figure 5.9 Pressure surges upon closing LBV-3 and LBV-4......................................................59

Figure 5.10 Pressure increase in well branch upon closing well choke .....................................60

Figure 5.11 Schematic of heat transfer assumptions used in Olga models................................62

Figure 5.12 Effective heat transfer coefficient for the different soil descriptions at steady state .63

Figure 5.13 Prediction of pipe wall temperature for different heat transfer assumptions ............64

Figure 5.14 Pipe wall temperature given different initial conditions............................................65

Figure 5.15 Temperatures during venting and relation to the CO2 phase boundary...................66

Figure 5.16 Impact of leak size on transient pipe wall temperature ...........................................67

Figure 5.17 Impact of leak size on minimum pipe wall temperature...........................................68



Rev 01

Heavy Oil

07-2-LA-5507-0003

Figure 5.18 Impact of flowline length on pipe wall temperature .................................................69

Figure 5.19 Comparison between simplified and detailed model for a 4” vent ...........................70

Figure 5.20 Comparison between a 4” and 6” vent for the detailed model.................................71

Figure 5.21 Flowline temperature profiles during venting process (Scotford to Well 1) ..............71

Figure 5.22 Vent rate for a 4” vent.............................................................................................72

Figure 5.23 Vent rate for a 6” vent.............................................................................................72

Figure 5.24 Impact of soil properties and ambient temperature on pipe wall temperature .........73

Figure 5.25 Vent rate for 4” and 6” ............................................................................................74

Figure 5.26 Temperature/Pressure conditions during venting and relation to phase envelope ..75

Figure 5.27 Minimum temperatures during venting of section between LBV3 and LBV4 ...........76

Figure 5.28 Holdup at minimum temperature during venting between LBV3 and LBV4 .............77

Figure 5.29 Liquid level in well during blowdown (initial time) ....................................................79

Figure 5.30 Liquid level in well during blowdown (steady state) ................................................79

Figure 5.31 Temperature isotherms in well during blowdown (steady state)..............................80

Figure 5.32 Temperature isotherms in well during blowdown (initial time) .................................80

Figure 5.33 Safety valve setting based on single phase and hydrate criteria.............................81

Figure 5.34 Impact of reservoir injectivity on safety valve setting ..............................................82

List of Tables

Table 3.1 Summary of reservoir characteristics (3) ...................................................................13

Table 3.2 Fluid composition of injection fluid (3)........................................................................14

Table 3.3 Flowline and Branch details (3) .................................................................................21

Table 3.4 Summary of pipeline operating conditions (3) ............................................................22

Table 4.1 Summary of CO2 rates to avoid hydrate region..........................................................31

Table 4.2 Summary of winter operation operating envelope ......................................................45

Table 4.3 Summary of summer operation operating envelope ..................................................46

Table 4.4 Summary of number of injection wells required .........................................................46

Table 5.1 Summary of fluid hammer cases for the wellbore injection scenario ..........................48

Table 5.2 Material physical properties used in Olga modeling ...................................................62

Table 5.3 Summary of valve closing time base on single phase criteria in well .........................81



Rev 01

Heavy Oil

07-2-LA-5507-0003

SUMMARY

This work outlines the flow assurance recommendations for the Quest project. The main flow assurance issues expected are due to hydrates and cold temperatures. In both cases, these issues can be mitigated by chemicals and/or operating procedures. Although it should be noted that a conservative hydrate strategy was adopted and there is reasonable confidence that hydrates formation is a very low risk in this system.

Based on the work performed in this study, several areas deserve additional study. These areas mainly focus on the ability to adequately model the system with the tools currently available. The thermodynamics used in steady-state calculations and for the hydrates curve are reasonably robust, but the transient simulations have fundamental limitations. The OLGA dynamic simulator cannot treat impurities or the solid state properly. The impurities will move the phase envelop to higher pressures, for example. Other transient conditions such as startup, ramp-up or ramp-down requires a thorough understanding due to the non-intuitive behavior shown already in this study. Due to vaporization, condensation and a high coefficient of expansion of the dense phase, the time temperature and pressure history can be misinterpreted which could lead to unwarranted actions.

However, that being said, the recommendations for additional hydrate data are highlighted for technical completeness. These data will not have a material impact on the overall feasibility of this system as a conservative mitigation strategy was recommended.

As part of this work, several recommendations were made.

A flowline vent size of 4” or less is recommended o Larger vent sizes result in the fluid temperature in the flowline to decrease below

the material integrity limits of the pipe.

A surface controlled subsurface safety valve (SC-SSSV) setting of 1,000 m is recommended

o Depth based on two criteria Maintaining a single liquid phase across the valve Preventing pressure surges from exceeding the maximum bottomhole

pressure o Depth can be shallower, if valve can be assured of closing quickly

Validate hydrate equilibrium data in single phase Liquid CO2 region at low water content o Current hydrate strategy is based on most conservative predictions o Potential to use a less conservative hydrate mitigation strategy

Define detailed operating procedures based on final design o Needed for flowline venting o Needed for initial line fill o Needed for initial displacement of wells



Rev 01

Heavy Oil

07-2-LA-5507-0003

1.0 BACKGROUND

The Quest Carbon Capture and Storage (CCS) project transports CO2 from the Scotford

upgrader in Alberta, Canada to an underground aquifer. This work is aimed at highlighting potential flow assurance risks associated with this project.

The design capacity of the system is to be able to capture 1.2 Mtpa of CO2. To achieve this, a 12” flowline is routed from Scotford to a series of injection wells. The furthest injection well is located about 84 km from the Scotford upgrader.

Flow assurance for Quest prospect during Conceptual, Selection and Define design phases has considered the following 5 main aspects:

Design of the Surface System (e.g. Pipeline, Valves, Wellbore)

o Thermal-hydraulic performance of the system

o CO2 Pipeline sizing and compressor requirements

o Maximum system capacity

o Insulation Requirements

o Vent-valve design

o Design requirements for above ground section of pipelines

Operability of the System

o Operability for normal operation

o Low flow events

o Emergency pipeline Leak/Blowdown

o Emergency wellbore blowout

o System start-up

o Vent-line operability

o Liquid hammer impact

o Low-water content operability

Solids Deposition Risk: Hydrates

o Dehydration limits

o Mitigation options

Multiphase Flow Aspects

o Two-phase flow in pipeline and wellbores

o Slugging potential

o Liquid hammer



Rev 01

Heavy Oil

07-2-LA-5507-0003

• Modeling Aspects

o Simulators applicability

o Impurities Impact



Rev 01

Heavy Oil

07-2-LA-5507-0003

2.0 FLOW ASSURANCE STRATEGIES

This section describes the flow assurance strategies used to mitigate each of the flow assurance risks. In addition, the flow assurance strategies associated with the main operating modes (start-up, steady state, and shut-in) were also addressed.

2.1 Solids Management

2.1.1 Hydrates

Hydrates will be managed primarily by dehydration of the injection fluids to sufficiently remove water to inhibit the formation of hydrates. Hydrate formation downstream of the well choke is not expected. However, if hydrates do form, they will be managed by chemical (methanol) injection. In exposed sections of the pipeline (low ambient temperatures) hydrate formation will not be mitigated, but provisions will be made to remediate any hydrates formation should it be required.

2.1.2 Wax

The injection fluid does not contain any wax.

2.1.3 Pour point

The injection fluid does not have any associated pour point issues.

2.1.4 Asphaltenes

The injection fluid does not contain any asphaltene.

2.1.5 Scale

Scale formation will be mitigated by dehydration of the injection fluid.

2.1.6 Corrosion

Corrosion of the flowline will be managed by ensure that the injection fluids are sufficiently dehydrated.



Rev 01

Heavy Oil

07-2-LA-5507-0003

2.1.7 Emulsions

Emulsions are not expected to be an issue.

2.1.8 Slugging

Slugging is not an issue with these fluids. The proposed operating conditions require that the fluid be in the single phase region. A previous study (1) looked at potential operation in two- phase flow and did not identify any slugging behavior.

2.1.9 Injected Solids

To prevent any reservoir impairment due to injected solids, a filter will be installed at each wellhead (2).

2.1.10 Chilly Choke

There is a large pressure drop taken across the well choke which results in some Joule- Thomson cooling. At typical operating conditions, the temperatures observed are well above any material integrity limits. Hydrate/ice formation may be an issue, but will be mitigated, as described above. Based on typical operating conditions, the lowest temperature expected downstream of the well choke is about - -10°C, which is not sufficiently low to cause any issues.

There is considerable cooling anticipated during a blowdown of the flowline. The vent pipe will be constructed of a material that can handle the low temperatures. Low temperatures in the flowline will need to be managed by correctly implemented operating procedures, to be defined during detailed design.

2.2 Operational Considerations

2.2.1 Start-Up

Initial startup of the flowline could be the most problematic. Based on a previous study (1), the initial line fill shows non-intuitive behavior, in that the pressure does not systematically increase with increasing amounts of CO2 injected to the flowline. The CO2 condenses as the pressure is

increased and the liquid phase is highly compressible. As a result, the initial startup needs to be carefully modeled, to accurately incorporate the startup procedure, i.e. the compressor output as a function of time.



Rev 01

Heavy Oil

07-2-LA-5507-0003

2.2.2 Steady-state

There should not be any flow assurance related issues during normal steady state production.

2.2.3 Shut-In

The injected fluid is sufficiently dehydrated so that there should be not issues upon shut-in. The only area of concern may the exposed section of pipe at the line break valves and the well pads. Upon shut-in, there is insufficient water to form any type of blockage and upon restart and deposits that were formed would be easily removed as the flow warms those bare sections of pipe.

Fluid hammer was evaluated during a shut-in. In all cases, the pressure surge in the system was less than the maximum system design pressure of 147.9 bar. Injection of the full design rate of 1.2 Mtpa into well 1 while the system is operating at the maximum design pressure resulted in pressure surges that were close to the maximum design rating of the pipeline. At the normal operating pressures of 120 bar, the pressure surges predicted were all much less than the maximum design pressure in the pipeline.

2.2.4 Flowline Venting

The venting of the flowline will need to be completed properly to ensure that cold temperatures are not observed in the flowline. Once a better definition of how the system pressure will be reduced, further simulation work is recommended to determine the detailed operating procedures.



Rev 01

Heavy Oil

07-2-LA-5507-0003

3.0 BASIC DATA

3.1 PVT and Reservoir Data

A summary of the reservoir data is shown in Table 3.1. There is a considerable range of reservoir injectivity values considered, so most simulations include some sensitivity to this value, with the worst case being either the low or high value depending on the particular scenario. Figure 3.1 shows the original geothermal gradient used in the wellbore modeling. Note the reservoir temperatures do not match between Figure 3.1 and Table 3.1. The DTS trace is believed to have been taken prior to establishing thermal equilibrium, so the linear approximation is probably more accurate. This does not have any impact on any of the flowline modeling or wellbore injection scenarios. The only time this difference in reservoir temperature will have an impact is on the modeling of the wellbore blowout.

Table 3.1 Summary of reservoir characteristics (3)

Reservoir Temperature [degC] 60

Reservoir Pressure [bar] 200

Max allowable bottomhole pressure [bar] 280

Reservoir Injectivity [Mm3/d/bar]

Low 8,665

Base 22,800

High 349,000



Rev 01

Heavy Oil

07-2-LA-5507-0003

Component Normal Operation

Mole%

Upset Condition

Mole%

CO2 99.2 95

CO .02 .15

N2 0 .01

H2 .68 4.27

Methane .09 .57

Water <52 ppm 52 ppm

Figure 3.1 Geothermal temperature profile for wells (4)

3.2 Fluid Compositions

Fluid compositions are defined in Table 3.2 for both the normal and upset cases. In all OLGA simulations, due to limitations in the model, a pure CO2 stream was used.

Table 3.2 Fluid composition of injection fluid (3)



Rev 01

Heavy Oil

07-2-LA-5507-0003

3.3 Water Samples

Free water is not expected in any injection scenario. The design case is to dehydrate the CO2

to less than 6 lbs/MMscf during injection. The performance of the TEG unit used to dehydrate the CO2 is a function of the ambient temperature. During normal winter operations, i.e. colder

ambient temperature, the water content should be about 4 lbs/MMscf, while at the warmer summer temperatures, the water content is increased to about 6 lbs/MMscf. During both the anticipated winter and summer operations, all water is dissolved in the CO2 phase.

3.4 Hydrates

It is possible to form hydrates from mixtures containing CO2 and water. Figure 3.3 shows the

predicted hydrate curve for the normal operating conditions. As previously reported (1), over the range of compositions expected between the normal and upset conditions, the impact to the hydrate equilibrium conditions is very small. The dehydration of the injected CO2 effectively

inhibits any hydrate formation during normal operating conditions in the pipeline. Based on the initial recommendations, dehydration of the injected CO2 was sufficient to prevent hydrates at

normal shut-in conditions of 0°C and 140 bar. However, hydrate formation was still possible during events, such as JT cooling across the well choke.

As part of this work, the validation of hydrate equilibrium in the presence of a small amount of water was investigated more closely. Figure 3.2 shows a comparison of the STFlash (in-house Shell software) and MultiFlash (commercial software) and how well they predict the water content of liquid CO2 near the region of interest for the Quest project. Note that STFlash

matched the data quite well, while MultiFlash under-predicted the data by an order of magnitude. The data and STFlash show that water is quite soluble in liquid CO2. This implies

that hydrate formation in the presence of liquid CO2 is inhibited because the water is highly

soluble in the liquid CO2.

Figure 3.3 shows an update hydrate equilibrium curve for both STFlash and MultiFlash. In the presence of free water, both programs predict nearly the same hydrate equilibrium curve. As the water content is decreased, the two programs begin to diverge in their predictions. The figure shows that the increase water solubility predicted by STFlash is sufficient to prevent the formation of hydrates. Conversely, MultiFlash predicts hydrates are stable even in the presence of liquid CO2. Note that this difference only occurs at low water content with liquid CO2. At all

other conditions, the two programs predict very similar results.

Based on the data, the STFlash predictions would appear to be more accurate. However, note that there is very limited data available to benchmark the models. And the data themselves are difficult to measure and prone to errors. Therefore, to be conservative, the MultiFlash predictions are still being used in developing the hydrate mitigation strategies, but it is recognized that this may be overly conservative.



Rev 01

Heavy Oil

07-2-LA-5507-0003

Figure 3.2 Model prediction and comparison with experimental data (5)

Figure 3.3 Predicted hydrate curve for composition during normal operation

3.4.1 Hydrate Inhibition Requirements

In several of the cases, it is possible to form hydrates, which means that a hydrate mitigation strategy is needed. One option is to use a hydrate inhibitor to prevent the formation of hydrates. In this section, the dosage requirements for the prevention of hydrates using methanol are given. The results are given in Figure 3.4 and Figure 3.5. The first figure shows the impact of temperature at a constant pressure. As is typical, the lower the temperature, the higher the



Rev 01

Heavy Oil

07-2-LA-5507-0003

methanol requirement to fully prevent hydrate formation. The methanol dosage requirement continues to decrease with increasing temperature until the temperature is sufficiently high that hydrates are no longer stable. The second figure shows similar results, but here as the pressure was decreased, the methanol requirement is shown to decrease as well.

In both figures, the methanol requirement is shown as a function of the water content. For all conditions given, the higher the water content, the higher the methanol dosage requirement required to prevent hydrates. Also in both cases, there was a sharp break in the curve predicted, which was the result of the differing water content of the fluid.

As with the prediction of the hydrate curve, methanol solubility in the liquid CO2 phase is difficult

to predict. Based on limited data, the actual methanol values may be twice as high as given in the figures. Despite these high methanol dosage rates, given that the water content is low (<50 ppm), the total methanol volume requirements will also be low. However, it is recommended to experimentally validate the hydrate curve and methanol requirements to allow for a less conservative design in the future.

Figure 3.4 Methanol requirement for hydrate inhibition – Impact of temperature



Rev 01

Heavy Oil

07-2-LA-5507-0003

Figure 3.5 Methanol requirement for hydrate inhibition – Impact of water content

3.5 Chilly Choke

The Joule-Thomson curves are presented in this section. Based on the expected well choke pressure drop, the temperature decrease was not sufficient to present any materials issues. However, hydrate formation could be an issue. The JT cooling curves are presented along with the MultiFlash predicted hydrate phase boundary (Figure 3.6). Hydrate curves are given on each figure to correspond to the section below which hydrates are stable. At 6 lbs/MMscf, the hydrate formation conditions were more severe, i.e. at a given pressure the hydrates form at a higher temperature. Fortunately, during summer operation, when the higher water content is expected, the temperatures are also higher. The figure details how much pressure drop is allowed at the wellhead before hydrate formation is possible. At the higher water content, a pressure d/s of the choke of about 45 bar is required before hydrates are stable. During winter operation, the temperatures are colder, which means that less pressure drop is allowed before hydrates can form.

Figure 3.7 shows a similar plot, but instead using the STFlash generated hydrate curve. There is a very marked difference in the region of hydrate stability. In these cases, the pressure downstream of the choke needs to be reduced to about 25 bar before hydrates are stable. Given that the STFlash predictions are probably more accurate than the Multiflash predicted values, the injection wells can probably operate without continuous hydrate inhibition. However, the formation of hydrates cannot be sufficiently discounted, so it is still recommended to have a hydrate mitigation strategy in place.



Rev 01

Heavy Oil

07-2-LA-5507-0003

Figure 3.6 Hydrate formation in relation to JT cooling across well choke (MultiFlash)



Rev 01

Heavy Oil

07-2-LA-5507-0003

Figure 3.7 Hydrate formation in relation to JT cooling across well choke (STFlash)

3.6 Wax and Pour Point

There are no wax are pour point issues with the injection fluid.

3.7 Asphaltenes

There are no asphaltene components in the injection fluid.

3.8 Scale

Scale is not expected to be an issue. The injection fluid is sufficiently dehydrated that no free water exists in the system.

3.9 Well Details

Details of the well geometry are unknown at this time. In all cases, it was assumed that the well was a 2000 m vertical pipe. The well ID was assumed to be 100.5 mm (3.957”). In the steady

state models, a constant heat transfer coefficient of 11.36 W/m2-K was used. In the transient simulations, a soil layer was included in the model.



Rev 01

Heavy Oil

07-2-LA-5507-0003

3.10 Flowline Details

The system was modeled per the details in Table 3.3 and Figure 3.8. The well branches were not defined in detail. In this work, it was assumed that the well branches consisted of a 5km horizontal line at each of the well locations indicated on the flowline topography.

Each of the Line Break Valves (LBVs) was included in the model along with the associated above ground section lengths as defined in Table 3.3. These sections were modeled using the given ambient temperature conditions and assuming a heat transfer coefficient consistent with a bare pipe. The well pads were modeled similarly as a section located at the end of each of the 5 km well branches.

Table 3.3 Flowline and Branch details (3)

Main Trunk Line Well Branches

Diameter-OD [mm] 323.9 168.3

Diameter-ID [mm] 299.7 146.3

Wall Thickness [mm] 12.1 11

Length [km] 80.4 5

Minimum Burial Depth [m] 1.5 1.5

Average above ground length at LBV or well pads

20 25



Rev 01

Heavy Oil

07-2-LA-5507-0003

Figure 3.8 Detailed flowline topography with location of LBVs and well branches

Table 3.4 Summary of pipeline operating conditions (3)

Winter Conditions

Summer Conditions

Pipeline Inlet Temperature [degC] 43 49

Operating Pressure [barg]

Normal Min 80 80

Normal Max 110 110

Maximum Design 140 140

Flow rate [Mtpa]

Minimum 0 0

Expected 1.2 1.2

Ambient Temperature [degC] -40 35



Rev 01

Heavy Oil

07-2-LA-5507-0003

Ground Temperature at flowline burial depth [degC] 0 11

Heat Transfer Coefficient [BTU/hr-ft2-F]

Minimum 0.35 0.35

Maximum 1.0 1.0

3.11 Production Function

A production function was not assumed in this work. The CO2 was assumed to be injected at a rate of somewhere between the extremes defined in

Table 3.4. Where appropriate, a range of flow rates were used to capture the complete operating envelope. The operating procedures have not been developed as to define injection rates into individual wells.



Rev 01

Heavy Oil

07-2-LA-5507-0003

4.0 STEADY STATE ANALYSIS

Steady state analysis for all simulations was completed using Unisim Design R390 using the shell flow correlations v5. A detailed topography was using in Unisim as per Figure 3.8.

4.1 Low Flow Events

This study addresses potential flow assurance issues due to low flow in the Quest pipeline and subsequent injection into the well. This low flow scenario was envisaged when a substantial

amount of the CO2 is diverted away from the injection wells into a 3rd party pipeline. This work

aimed to determine the minimum flow rate required in the Quest system and the resulting flow assurance issues. The flow assurance risks identified for a low flow scenario included potential for significant JT cooling across the well choke, which could result in hydrate formation. This was in part addressed in the chilly choke section, but is elaborated on here with specifics of the injection case. There is also the potential to under or over pressure the line if a sufficient inventory of CO2 is not maintained in the flowline.

4.1.1 Wellbore Flowing Conditions

In these simulations, the flowing wellhead pressure (downstream of the choke) was important to determine. In order to determine this value, the flowing bottomhole pressure is also required. The FBHP can be determined given the reservoir pressure and the reservoir injectivity values. These results are shown in Figure 4.1. Also given that there is a maximum FBHP, this figure can be used to determine the maximum possible injection rate into a well.

These FBHP were then used to determine what the flowing pressure at the wellhead was just downstream of the well choke (Figure 4.2). An interesting feature to these curves was that there is a decreasing requirement in well pressure at low rates and then an increase as the rates increase. The decrease in pressure is attributed to the CO2 density (Figure 4.3) increase

with decreasing temperature. At low rates, the fluid temperature in the well was near the geothermal gradient. As fluid rates increased, an increasing amount of cold fluid was flowed into the well and an increasing amount of the well was filled with colder fluids.

Figure 4.4 shows the temperature profile in the well at several of the lower injections rates. At the lowest rates, there was a step-wise appearance to the temperature, which was due to the discretization of the wellbore in Unisim and the application of a constant ambient temperature to a given section. This effect can be minimized by using a smaller segment size in the well. As the rates increased, this effect disappears. So the figure is probably not extremely accurate at the low rates, but the general trend of a decreasing temperature in the wellbore is accurate.



Rev 01

Heavy Oil

07-2-LA-5507-0003

The lower temperatures in the well, at the higher rates, result in an increase in the CO2 density. The increase in density results in a larger hydrostatic pressure drop along the well. Over the 2,000 m length of the well, this slight difference in fluid density results in a noticeable change in the wellhead pressure. At higher rates, the FBHP increases sufficiently and frictional pressure drop begins to become important at which time the wellhead pressure begins to increase. This meant that the decrease in wellhead pressure at low rates was a real phenomena and results in a stable flow regime.

Figure 4.1 Flowing bottomhole pressure as a function of CO2 injection rate



Rev 01

Heavy Oil

07-2-LA-5507-0003

Figure 4.2 Wellhead pressure (downstream of choke) as a function of the CO2 rate

Figure 4.3 CO2 density as a function of temperature



Rev 01

Heavy Oil

07-2-LA-5507-0003

Figure 4.4 Wellbore temperature at different CO2 injection rates with hydrate curve

4.1.2 Hydrate Risk

Based on the Shell in-house software, the risk of hydrate formation is very low. However, due to uncertainties in current hydrate equilibrium predictions, the hydrate strategy is based on the more conservative predictions given using Multiflash.

Over a fairly wide range of rates, it is possible to form hydrates in the wellbore. Figure 4.5 shows the pressure at which hydrates begin to form along with the well injection curve. When the injection curve is above the hydrate stability boundary, no hydrates are formed. Thus, the injection rate needs to be high enough to raise the pressure downstream of the choke above this boundary. Due to injection curve first decreasing and then increasing as explained previously, the injection curve can cross the hydrate stability curve twice meaning that very low or high injection rates are feasible without hydrate formation. Finally, too high of injectivity will mean additional hydrate mitigation will be required to have a robust operating envelope.



Rev 01

Heavy Oil

07-2-LA-5507-0003

Figure 4.5 Hydrate risk at 4 lbs H2O/MMscf water content

Figure 4.6 Hydrate risk at 6 lbs H2O/MMscf water content

Figure 4.6 shows a similar result, but for CO2 injection with a water content of 6 lbs/MMscf. In this case, the pressures at the wellhead are all relatively lower before hydrates can form.



Rev 01

Heavy Oil

07-2-LA-5507-0003

Roughly, the 11°C results at 6 lbs H2O/MMscf at all wellhead pressures were similar to the 0°C

results at 4 lbs H2O/MMscf for the 8 MPa pressure. Note that the 0°C case with

6 lbs H2O/MMscf is not shown because the entire system, including the flowline, operated in the hydrate region was not considered a feasible scenario.

4.1.3 Hydrate Mitigation

Based on the assumption of the conservative hydrate predictions, some form of hydrate mitigation is required. Mitigation methods for hydrate prevention include increasing the temperature, reducing the pressure, adding inhibitors, or reducing the water content of the CO2

further.

In reducing the pressure, the intent is to reduce the pressure drop across the well choke. For a given PI and injection rate, the wellhead pressure downstream of the choke is fixed. To reduce the pressure drop, one option is to lower the flowline pressure. Alternatively, the well choke could be moved downhole, which would minimize the pressure drop across the choke. Another method is to decrease the reservoir PI, which increases the wellhead pressure downstream of the choke. Unless there is a requirement that one well be able to take the full flow rate of CO2,

there is not a compelling reason to have a low reservoir PI. However, short of purposely damaging the formation there is minimal control over this value. To minimize the hydrate risk, the flowline should be operated at as low of a pressure as practical. However, this alone does not eliminate the hydrate risk.

The addition of heat to the system is probably impractical, given that the hydrate risk occurs in the wellbore over a relatively large length of the tubing. The predicted amount of methanol required to inhibit hydrates is about 1-2 times the water content (on a mass basis). There is a lot of uncertainty in this calculation, so this value could be off by +/- 100%. This means that the injection rates are likely to be large for any significant flow rates into the well. Furthermore, decreasing the water content of the CO2, while eliminating the hydrate risk, is probably difficult

to implement. However, if any hydrate risk in the system is deemed unacceptable, this option should be explored further.

The best option for hydrate mitigation may be to do nothing. Given the small amount of water in the CO2, the buildup of hydrates is slow. Based on the thermal modeling, only a small portion of

the wellbore exists in the hydrate region, so any hydrate that is formed is more likely to be carried along with the cold fluids than it is to stick to the relatively warmer pipe wall. At these conditions it is only possible to convert 20%-50% of the total water to hydrate. And given that at shut-in conditions, the entire well is outside the hydrate region, a better hydrate strategy may be to periodically switch the injection wells and allow any hydrates in the well to melt naturally. In this strategy, some heat tracing of the exposed section of the wellbore downstream of the choke is required to ensure that all of hydrates do indeed melt.



Rev 01

Heavy Oil

07-2-LA-5507-0003

However, in allowing hydrates to form, there is the possibility of creating free water in the system as the hydrates melt. This free water could present a corrosion risk in the wellbore.

4.1.4 Flow Assurance Risks in Pipeline

Based on this analysis there is minimal flow assurance risk in the pipeline. The biggest concern is likely to be keeping the pipeline in the single phase region.

There is a potential for pressure fluctuations in the pipeline due to the temperature variations expected with the changing seasons. If there is no flow into the pipeline, the mass of CO2 in the

pipeline and the pipeline volume are fixed. Since the density is simply the mass divided by the volume, both of which are constant, this means that the pressure must change to keep the fluid density in the pipeline constant. As the temperature is increased (winter to summer) the CO2

density tries to decrease (Figure 4.3), which means that system pressure must increase to keep

the fluid density constant. In changing the temperature from 0°C to 11°C, it is possible to increase the flowline from 10 MPa to about 16 MPa due to this effect. Conversely, it is possible to decrease the flowline pressure by about 6 MPa in decreasing the temperature in going from summer to winter. This decrease in pressure is sufficient to move the flowline into the two- phase region.

To mitigate this problem, there needs to be some flow into or out of the system, depending on the change in temperature. In going from summer to winter, CO2 needs to be added to the pipeline while in going from winter to summer CO2 needs to be removed (injected into the wells).

The mass of CO2 needed to maintain a constant pressure represents about 10% of the pipeline

volume (percent change in CO2 density over this temperature range). Assuming a 6 month time frame to add this CO2 to the pipeline, this corresponds to an injection rate of about 0.01 Mtpa. The rate could be higher if the duration for injection is shorter. The rate can also be managed by the high and low pressure automated alarms and controls (venting at compressor and closing of the well choke respectively).

4.1.5 Conclusions

Based on this assessment, there is some risk associated with a low flow scenario due to hydrate formation downstream of the well choke. Based on this hydrate risk, the minimum flow rates in the flowline/wells can be determined. The single phase criterion in the flowline is used to determine the minimum flow rate required.

Table 4.1 summarizes the results of this analysis. Because of the nature of the flow curves, there is a region at very low flow rates that is outside the hydrate region and an area at high rates that is outside the hydrate region. It is only at intermediate rates between these values where hydrate formation is possible. This region of hydrate risk depends strongly on the flowline conditions (temperature, pressure, and water content) and the reservoir injectivity.



Rev 01

Heavy Oil

07-2-LA-5507-0003

Considering the conservative hydrate criteria, the minimum flow in the well should not exceed about 0.02 Mtpa in the low flow regime. Similarly, in order to ensure that enough CO2 is

available to maintain pressure in the line, about 0.01 Mtpa is required. So at the low end of the flow curve, maintaining flowline pressure in the single phase region (due to changing ambient temperature) indicates that some availability of CO2 is required at a rate of about 0.01 Mtpa.

If this regime is not practical, then much higher rates are required to completely avoid any hydrate risk. The recommendation in this case is to maintain the flowline at as low of a pressure as is practical. But even in the most optimistic of cases, the required flow rate is greater than 0.3 Mtpa and more likely about 0.54 Mtpa for the base case assumptions.

At flow rates in the range of 0.01 to 0.54 Mtpa, some hydrate risk will be present. Although this risk is low, so it is likely that the system could be operated by periodically switching from one injection well to another. More detailed work needs to be done to determine this frequency but it is likely to be on the order of weeks to months and not days. Some work would need to be done to quantify the corrosion risk due to the presence of free water during hydrate dissociation. If no hydrate risk is permissible, then other options, such as methanol injection or decreasing the water content of the CO2 is required.

Table 4.1 Summary of CO2 rates to avoid hydrate region

WellHead U/S Choke Low PI Base PI High PI

Temp. Press. Water Rate < Rate > Rate < Rate > Rate < Rate >

[C] [Mpa] [lbs/MM] [Mtpa] [Mtpa] [Mtpa] [Mtpa] [Mtpa] [Mtpa]

0 8 4 0.02 0.11 0.02 0.33 0.02 0.92

0 10 4 --- 0.23 --- 0.54 --- ---

0 14 4 --- 0.42 --- 0.88 --- ---

11 8 4 --- 0 --- 0 0.09 0.52

11 10 4 --- 0 --- 0 0.06 0.60

11 14 4 --- 0 --- 0 0.06 0.63

0 8 6 --- --- --- --- --- ---

0 10 6 --- --- --- --- --- ---

0 14 6 --- --- --- --- --- ---

11 8 6 --- 0 0.03 0.26 0.02 0.81

11 10 6 0.03 0.03 0.02 0.31 0.02 0.89

11 14 6 0.01 0.13 0.01 0.38 0.02 0.99

4.2 Normal Production – Temperatures for Above-Ground Piping Sections



Rev 01

Heavy Oil

07-2-LA-5507-0003

The steady state temperatures are presented at all the relevant line break valve (Figure 4.7 through Figure 4.13) and well pad locations (Figure 4.15 and Figure 4.16). All these locations corresponded to the above ground sections in the system.

For all the LBV locations, the temperatures showed a similar trend to one another. At very low rates, the temperature was close to the ambient temperature. As the rate was increased, the temperatures became closer to the compressor discharge (line inlet) temperature at locations close the Scotford end of the flowline. At locations further away from Scotford, temperatures were much closer to the ambient temperature at the line burial depth.

Figure 4.14 shows the sensitivity of the heat transfer coefficient on the temperature at LBV-1. As the heat transfer coefficient was increased, a relatively larger amount of heat was lost to the surrounding soil. Hence the temperature at LBV decreased as the heat transfer coefficient was increased. Temperatures at LBV-1 were always highest, due to closer proximity to the warmer inlet conditions. As the distance from the Scotford end was increased, the heat transfer coefficient had less and less impact since the fluid temperatures were close the ambient soil temperature.

Similar results were observed for all the well pad locations (Figure 4.15 and Figure 4.16), but since they were located relatively far from the Scotford end, they were most similar to the LBV-7 temperatures. Only well pads 1 and 5 are shown in this analysis as these two locations represent the two bounds for all the well pad locations, so any temperature at the other well pad locations is somewhere between these two extremes, which already show minimal variation in the temperatures.

4.2.1 Figures for Steady State Temperatures at all Line Break Valves



Rev 01

Heavy Oil

07-2-LA-5507-0003

Figure 4.7 Temperature at LBV-1 during steady state operation




Rev 01

Heavy Oil

07-2-LA-5507-0003




Rev 01

Heavy Oil

07-2-LA-5507-0003





Rev 01

Heavy Oil

07-2-LA-5507-0003

4.2.1.1 Temperature sensitivity to heat transfer coefficient at LBV-1

Figure 4.14 Impact of heat transfer coefficient on the temperature at LBV-1

4.2.2 Figures for well pad locations 1 and 5



Rev 01

Heavy Oil

07-2-LA-5507-0003

Figure 4.15 Arrival temperature at Well 1



Rev 01

Heavy Oil

07-2-LA-5507-0003

Figure 4.16 Arrival temperature at Well 5

4.3 Normal Production – Flowline and Well Operating Envelopes

Figure 4.17 shows a simplified schematic of the operating lines upstream and downstream of the well choke. The ‘Flowline Operating Lines’ take into account the pressure drop in the flowline and the well branches and represents the pressure that would be expected upstream of the well choke at a given flow rate. In this case it is assumed that all flow is routed from the Scotford end to a particular well. In this work, wells 1 and 5 were again chosen as they represent the range of operating conditions that can be expected. The ‘Well Operating Line’ was determined in the previous section. Again it takes into account the pressure drop along the wellbore and incorporated the FBHP requirement as determined by the reservoir injectivity. This pressure represents the pressure downstream of the choke required to flow a given rate into the reservoir.

The difference between these two lines represents the pressure drop that would be taken at the wellhead to achieve the given flow. The rate where these two lines cross represents where there would be no pressure drop at the choke. At any injection rate higher than this, there would not be sufficient pressure at the wellhead to achieve the desired injection rate into the formation. There is also an additional constraint based on the maximum FBHP, which was



Rev 01

Heavy Oil

07-2-LA-5507-0003

assumed to be 280 bar in this work. If the FBHP constraint is met (i.e. it crosses the well operating line before it crosses the flowline operating line) then the maximum injection rate is constrained by the maximum FBHP, which means that there is a pressure drop across the well choke.

Figure 4.18 through Figure 4.21 show the operating charts for wells 1 and 5 with the normal and upset gas composition. These figures include the flowline operating lines for the range of expected flowline inlet pressures. The well operating lines cover the range of possible reservoir injectivity values. The pressure drop at the two well pad locations shows some variation with the different compositions and operating pressures. In the flowline at low pressure operation, there is sufficient pressure drop along the flowline such that the system enters into the two- phase region and gas is evolved. This gas evolution results in increased pressure drop, which results in more gas evolution. Hence the reason Figure 4.18 and Figure 4.20 show a large pressure decrease at high rates for the low pressure operations. Not surprisingly, the upset composition, which has a higher free gas content, shows a larger pressure drop for a given rate relative to the normal composition, as shown in Figure 4.19 and Figure 4.21.

Figure 4.17 Schematic for operating lines upstream and downstream of well choke



Rev 01

Heavy Oil

07-2-LA-5507-0003

Figure 4.18 Operating line for Well 1 with the normal composition



Rev 01

Heavy Oil

07-2-LA-5507-0003

Figure 4.19 Operating line for Well 1 with the upset composition



Rev 01

Heavy Oil

07-2-LA-5507-0003

Figure 4.20 Operating line for Well 5 with the normal composition

Figure 4.21 Operating line for Well 5 with the upset composition



Rev 01

Heavy Oil

07-2-LA-5507-0003

0.38 0.47 0.47

76 106 136

76 91 91

-1 -1 -1

-1 -2 -2

0.76 1.02 1.2

70 96 122

70 96 112

1 0 0

1 0 0

1.2 1.2 1.2

57 90 122

50 50 50

7 1 1

7 -1 -2

0.22 0.47 0.47

70 108 136

70 107 108

-1 -1 -1

-1 -1 -2

0.5 0.9 1.14

68 98 122

68 98 122

1 0 0

1 0 0

0.86 1.2 1.2

60 88 120

60 70 70

2 3 1

2 2 -2

0.37 0.47 0.47

75 104 134

75 91 91

-1 -1 -1

-1 -1 -2

0.74 1.01 1.2

68 93 118

68 93 112

-1 -1 -1

-1 -1 -1

1.2 1.2 1.2

54 86 118

50 50 50

-1 -1 -1

-1 -3 -4

0.19 0.46 0.47

66 114 134

66 114 115

-1 -1 -1

-1 -1 -2

0.44 0.87 1.12

65 96 119

65 96 119

-1 -1 -1

-1 -1 -1

0.76 1.2 1.2

58 85 116

58 70 70

-2 -1 -1

-2 -3 -4

Table 4.2 and Table 4.3 provide a summary of the operating conditions for a given well. Note that the cases are classified based on the constraint that determines the maximum injection rate.

‘No Constraint’, a single well was capable of injecting at the full design rate of 1.2 Mtpa.

‘Well PI constraint’ indicated that there was not sufficient wellhead pressure available to achieve the desired injection rate with the given well PI value.

‘BHP constraint’ indicated that a bottomhole pressure of 280 bar was exceeded with a higher injection rate.

This summary was also used to determine how many wells are required to inject the design rate of 1.2 Mtpa, as shown in Table 4.4. For most typical operations 1-3 wells were required. It should be noted that in the upset cases, more wells were typically required to inject a total of 1.2 Mtpa of CO2, which could cause operational issues if not recognized in time.

Table 4.2 Summary of winter operation operating envelope

No Constraint

Well PI Constrained

BHP Constrained

WINTER 80 bar 110 bar 140 bar 80 bar 110 bar 140 bar 80 bar 110 bar 140 bar

Well1-Norm Composition Max Rate [Mtpa]

WHP u/s choke [bar]

WHP d/s choke [bar]

WHT u/s choke [degC]

WHT d/s choke [degC]

Low PI Low PI Low PI Base PI Base PI Base PI High PI High PI High PI

Well1-Upset Composition Max Rate [Mtpa]

WHP u/s choke [bar]

WHP d/s choke [bar]




WHP u/s choke [bar]

WHP d/s choke [bar]




WHP u/s choke [bar]

WHP d/s choke [bar]





Rev 01

Heavy Oil

07-2-LA-5507-0003

0.38 0.47 0.47

76 106 136

76 91 91

11 11 11

11 10 7

0.76 1.02 1.2

70 96 122

70 96 112

14 13 13

14 13 12

1.17 1.2 1.2

49 90 122

49 50 50

12 14 13

12 12 8

0.24 0.47 0.47

72 108 136

72 107 108

11 -1 11

11 -1 10

0.5 0.88 1.12

68 96 120

68 96 120

11 13 13

11 13 13

0.8 1.2 1.2

59 88 120

59 70 70

11 14 13

11 12 8

0.37 0.47 0.47

75 104 134

75 91 91

11 11 11

11 10 9

0.74 1.01 1.2

68 93 118

68 93 112

11 11 11

11 11 11

1.2 1.2 1.2

50 86 118

50 50 50

11 11 11

11 9 8

0.37 0.47 0.47

69 114 134

69 114 115

11 11 11

11 11 9

0.74 1.01 1.2

66 96 118

66 96 118

10 11 11

10 11 11

1.2 1.2 1.2

58 85 116

58 70 70

8 11 11

8 10 9

Table 4.3 Summary of summer operation operating envelope

No Constraint

Well PI Constrained

BHP Constrained

SUMMER 80 bar 110 bar 140 bar 80 bar 110 bar 140 bar 80 bar 110 bar 140 bar


WHP u/s choke [bar]

WHP d/s choke [bar]





WHP u/s choke [bar]

WHP d/s choke [bar]




WHP u/s choke [bar]

WHP d/s choke [bar]




WHP u/s choke [bar]

WHP d/s choke [bar]



Table 4.4 Summary of number of injection wells required

80 bar 110 bar 140 bar 80 bar 110 bar 140 bar 80 bar 110 bar 140 bar


Winter

Well1-Norm Composition 4 3 3 2 2 1 1 1 1

Well1-Upset Composition 6 3 3 3 2 2 2 1 1



Summer







Rev 01

Heavy Oil

07-2-LA-5507-0003

5.0 TRANSIENT ANALYSIS

Transient modeling was completed with Olga v6.3 using the single component option for CO2

and with AFT Impulse version 4.0. Neither of these models included the impacts of impurities in the flow stream. However, given that the fluid properties of pure CO2 and the base composition

are very similar the impact on the results of the transient simulations was minimal. The only impact is due the base composition having a vapor-liquid region, whereas pure CO2 has a sharp

boundary between the liquid and vapor. However, most simulations take place in the single phase region thus this has minimum impact on the results.

5.1 Fluid Hammer

The impact of fluid hammer was investigated both in the wellbore and in the pipeline. The simulations consisted of closing the wellhead choke, the SC-SSSV, and the LBV. The SC- SSSV impacts both the minimum and maximum pressure observed at the formation, which may result in either back flow into the well or exceeding the formation fracture pressure. Similarly, the sudden closing of the wellhead choke or a LBV can result in a pressure surge in the pipeline that may exceed the design rating.

The fluid hammer simulations were completed using AFT Impulse and unlike OLGA, Impulse’s fluid hammer calculation is based on a liquid phase only. Furthermore, Impulse computes maximum theoretical pressure surge using the instantaneous water hammer equation which depends on the liquid density, the change in fluid velocity, and the wave speed.

Impulse does not contain CO2 as one of the standard components. But a ‘pseudo’ component

can be created in Impulse that mimics the fluid properties of CO2. The fluid properties required

to run Impulse are the vapor pressure, liquid density, viscosity and the bulk modulus of elasticity. Properties were obtained over the relevant temperature range from Unisim design and PVTsim. A limitation of the software was that the fluid properties are only a function of the temperature and not the pressure. The wave speed is a direct function of the density and bulk modulus. Because both properties also depend on pressure, the input values were chosen so that the resulting predictions would be conservative.

In addition, all fluid hammer simulations had the valve closing quickly, which in this work was 1 ms. The fast closing time was meant to represent worst case conditions. If any of the valves are closed more slowly, the magnitude of the pressure increase/decrease would be reduced. But in all cases, the maximum pressure surges observed were below any system limitations and

thus valve closing time should not be of concern. In order to correctly model the impact of closing the valve more slowly, details of the valve are required. Namely, how flow through the valve is impacted as the valve is closed.



Rev 01

Heavy Oil

07-2-LA-5507-0003

5.1.1 Fluid Hammer in the Wellbore

In the wellbore simulations, two scenarios were investigated. The first scenario looked at an injection case where the SC-SSSV was suddenly closed and the pressure surge above and below the valve was observed. In this case, the valve was assumed to be fully closed and did not allow the transfer of fluid across the valve. The actual valve is a flapper design, which allows fluid to be injected into the reservoir, but not from the reservoir to the wellhead. In the second set of cases, a well blowout was simulated in which the SC-SSSV was suddenly closed.

5.1.1.1 Wellbore Injection

Although an unlikely event, the wellbore injection case looked at a scenario where fluids were injected into the well and the SC-SSSV was closed suddenly. At a steady state injection rate of 1.2 Mtpa into a single well and given the base case assumption for reservoir injectivity, the BHP was about 270 bar. In this work, the maximum bottomhole pressure was constrained to 280 bar, which means that the SC-SSSV needs to be set at a depth that results in less than a 10 bar increase in bottomhole pressure during one of these type of shut-in events. Although this is an unrealistic case, it shows a number of sensitivities to illustrate the relative importance of the various parameters.

Table 5.1 summarizes the different cases investigated in these fluid hammer simulations. The base case was the injection of all CO2 (1.2 Mtpa) into a single well. The ‘Low PI’ case looks at

injecting the maximum rate of CO2 into a single well. Note in this case the maximum rate is

constrained by the maximum FBHP of 280 bar. The ‘Low Injection Rate’ case simulated a rate that represented the maximum pressure drop, across the well choke, and hence the coldest wellhead temperature. The ‘High Temperature’ case simulated the injection of CO2 into a single

well at the maximum rate during summer operation.

Table 5.1 Summary of fluid hammer cases for the wellbore injection scenario

Steady State

Scenarios Mass Q WHP WHT BHP

Mtpa bar C bar

Base Case 1.2 120 0 272

Low PI 0.4 86 0 274

Low Injection Rate 0.3 31 -5 227

High Temperature (Summer) 0.7 112 12 270

The base case resulted in the largest pressure surges. Figure 5.1 shows the maximum pressure expected upon closing the SC-SSSV while Figure 5.2 shows the minimum pressure.



Rev 01

Heavy Oil

07-2-LA-5507-0003

In the simulation the pressure surge travels, so all sections along the well do not experience the maximum or minimum pressure at the same time.

In Figure 5.1, the maximum pressure observed at the bottomhole was about 280 bar for the case where the SC-SSSV was set at 1,000 m. For any SC-SSSV setting deeper than 1,000 m, the maximum bottomhole pressure is likely to exceed 280 bar. This means that the SC-SSSV needs to be set at a depth shallower than 1,000 m in order to avoid the bottomhole pressure exceeding 280 bar. Similarly, the bottomhole pressure should not be less than the reservoir pressure in order to avoid the inflow of material from the reservoir upon shut-in. Based on these simulations, the bottomhole pressure is not expected to decrease below a reservoir pressure of 200 bar. This shows that as the valve is set further from the source, the expected pressure surge becomes larger, although the magnitude of the change is relatively small.

Figure 5.3 is similar to Figure 5.1, but shows the change in maximum pressure relative to the steady state value. In this case, the maximum pressure surge is always expected to occur just upstream of the SC-SSSV. The deeper the SC-SSSV is set, the larger in increase in surge pressure. Similarly, Figure 5.4 Figure 5.5 and Figure 5.6 show the expected pressure increases for the other cases investigated. Based on these results, the base case (i.e. highest rate and highest pressure) resulted in the largest surge pressures.

Note that the sharp pressure spikes at well depths of 1000 m and 1500 m are the results due to the vapor cavity forms and collapses as the conditions are near the vapor pressure curve.



Rev 01

Heavy Oil

07-2-LA-5507-0003

Figure 5.1 Maximum surge pressure in well upon shut-in



Rev 01

Heavy Oil

07-2-LA-5507-0003

Figure 5.2 Minimum surge pressure in well upon shut-in



Rev 01

Heavy Oil

07-2-LA-5507-0003

Figure 5.3 Increase in pressure above steady state values (Base Case)



Rev 01

Heavy Oil

07-2-LA-5507-0003

Figure 5.4 Increase in pressure above steady state values (low PI case)



Rev 01

Heavy Oil

07-2-LA-5507-0003

Figure 5.5 Increase in pressure above steady state values (Low Injection Rate)



Rev 01

Heavy Oil

07-2-LA-5507-0003

Figure 5.6 Increase in pressure above steady state values (High Temperature Case-Summer)

5.1.1.2 Well Blowout

The well blowout scenario simulated the case where there was an open path between the reservoir and the wellhead, which was at atmospheric pressure. The SC-SSSV was set at 1,000 m and is suddenly closed to prevent any further flow from the reservoir. The simulation of a well blowout used this SC-SSSV depth to investigate the magnitude of the pressure surges observed in the formation. Note that the bottomhole was open to the formation and the formation should be capable of absorbing the pressure surge. In the worst case, the formation injectivity is low enough that it is not capable of instantly absorbing the pressure surge, so the magnitude of the pressure surge should not exceed the fracture pressure.

Figure 5.7 shows the magnitude of pressure increase or decrease expected at the SC-SSSV. In the base case, the increase in pressure below the valve was about 20 bar. When this pressure surge is added to the reservoir pressure, the maximum pressure expected in the formation would be about 220 bar, which is well below the assumed fracture pressure of the formation. However, note that the figure shows there is no pressure increase expected at the



Rev 01

Heavy Oil

07-2-LA-5507-0003

bottomhole location due to the formation volume being modeled as a constant pressure boundary.

In order for the formation to be impacted by the surge pressure, the injectivity of the formation would have to be sufficiently low that the formation could not absorb the pressure surge quickly enough. In a low injectivity case, the flow rate in the well would be lower, which means if the SC-SSSV were suddenly closed, the pressure surge would be less than 20 bar, per the base case and the pressure at the formation would still be less than the fracture pressure.

Based on these results and the sensitivities completed for the injection case, a SC-SSSV setting of higher than 1,000 m will result in a higher pressure surge below the SC-SSSV. This would indicate that it is better to set the SC-SSSV deeper. Note that the magnitude of the pressure surges associated with the blowout case are less severe than the injection case. This is due to the lower starting pressure in the blowout case and also the fact that the blowout rates are less than the full design injection rate of 1.2 Mtpa. Based on these simulations, the pressure surge was not significant at the SC-SSSV, and therefore, there are no limitations on the closing time of this valve.

Figure 5.7 Change in pressure above and below SC-SSSV during blowout scenario



Rev 01

Heavy Oil

07-2-LA-5507-0003

5.1.2 Fluid Hammer in Flowline

It is also possible to experience pressure surges in the flowline upon closing a line break valve or a well choke. In these simulations, worst case conditions are observed at high pressure and low temperature, thus the pipeline inlet conditions assumed 140 bar and 0°C.

Several cases were simulated to observe the impact of closing one or more of the LBVs. Based on the simulations, the closing of LBV-1 represented the highest pressure surge. The case where the well choke was closed was also investigated. The big difference between the LBV case and the well choke case is the smaller diameter pipe leading to the well choke. The smaller diameter pipe results in larger fluid velocities, which result in greater pressure surges.

5.1.2.1 Closing of Line Break Valves

Figure 5.8 shows the scenario of closing LBV-1. In this simulation, the flowline was operated at the full design rate of 1.2 Mtpa just prior to the valve closing. The pressure increase observed at LBV-1 was about 3 bar. At a compressor discharge of 140 bar, the maximum pressure was less than the pipeline design pressure of 147 bar. Figure 5.9 shows a similar case, but for LBV-

3 closing. In this case the pressure increase was higher, at about 5 bar, but the absolute pressure was slightly lower due to the lower initial pressure at this location.

Similarly, Figure 5.9 shows the impact of closing two of the line break valves at the same time. LBV-3 and LBV-4 were chosen as they are the two valves closest to one another. The valves were both closed at the same time. This was meant to simulate the impact of trapped fluid in between the valves. The magnitude of the pressure surge was less for the fluids trapped between the two valves.



Rev 01

Heavy Oil

07-2-LA-5507-0003

Figure 5.8 Pressure surges upon closing LBV-1



Rev 01

Heavy Oil

07-2-LA-5507-0003

Figure 5.9 Pressure surges upon closing LBV-3 and LBV-4

5.1.2.2 Closing of Well Choke

The closing of the well choke represented the highest risk since the piping leading to the choke is of smaller diameter than the main trunk line. This smaller diameter results in a higher fluid velocity, which in turn means the pressure surge can be expected to be higher as well, since the two parameters are directly related.

The worst case was when all the CO2 was injected into well 1 at an operating pressure of 140 bar. Figure 5.10 shows the impact of the well branch length on the relative pressure increase when the valve is closed. For a 5.3 km well branch, the pressure increase is about

23 bar. In order to remain below the pressure limit of 147 bar, the wellhead pressure must be below about 124 bar. As per Figure 4.18, the pressure at wellhead 1, at the design rate of 1.2 Mtpa is about 122 bar. Note that Figure 4.18 through Figure 4.21 are only applicable for a 5 km well branch. This analysis shows that the pressure surges expected during this worst case operation are just on the boundary of system design.



Rev 01

Heavy Oil

07-2-LA-5507-0003

In order to ensure these pressure surges are not an issue due to the closing of the well choke, operation of the system at normal design pressure of 120 bar will ensure that the pressure surges are sufficiently low. If operation at high pressure is required, then multiple wells are required for CO2 injection.

Based on the fluid hammer simulations, the only potential issue due to pressure surges in the system was the closing of the well choke. Note that this was only a concern in the case of injecting the full design rate of 1.2 Mtpa into well 1 while the system was operated at the maximum design pressure of 140 bar. This issue can easily be mitigated by operating the system at the normal operating pressure of 120 bar. In the event that the high pressure is required, then two or more wells would be required for injection of the CO2. Additionally, the

well choke is expected to close more slowly than 1 ms, which will also lessen the magnitude of the pressure surge, but without detail information regarding the choke design, it was difficult to determine the actual impact.

Figure 5.10 Pressure increase in well branch upon closing well choke



Rev 01

Heavy Oil

07-2-LA-5507-0003

5.2 Flowline Venting

A number of different simulations were completed looking at the aspect of flowline venting. This was to model the impact of starting with a flowline initial filled with CO2 as a supercritical fluid

and then depressurizing the flowline to atmospheric pressure. The aim of these simulations was to determine what size of vent line would be required and to establish rough guidelines to be used in developing operating procedures during the venting process.

5.2.1 Description of heat transfer in model

Due to the transient nature of the venting process, the specifics associated with heat transfer become much more important.

Figure 5.11 shows a schematic of the heat transfer assumptions used in the models. In the steady state model, the ambient temperature was defined as the soil temperature at the flowline burial depth. An overall heat transfer coefficient was defined (as given in

Table 3.4) and that effectively defines the rate of heat transfer from the fluid to the surrounding soil. The first figure (with the 10mm) soil layer is representative of that steady state model whereby there is minimal thermal storage capacity of the soil and a steady state temperature profile is quickly developing from the pipe will to the defined ambient temperature. In order to model the transient nature of the venting process, the soil layer has to be included. The soil layer represents a large thermal mass that can and does change temperature during the venting process. Note that the soil layer is defined as 2.8 m instead of the given 1.5 m burial depth. This difference was due to the non-symmetric nature of the flowline burial and how it was accounted for in the modeling. In the actual case, the top of the pipe ‘sees’ 1.5 m of soil, but the bottom of the pipe ‘sees’ an infinite thickness of soil. In order to average out this impact, an effective soil layer thickness of 2.8 m was used per the hydraulic guidelines (6).

Figure 5.12 shows the Olga predicted heat transfer coefficient for these different assumptions. Note that there was not a large difference in the predicted values at steady state conditions. But Figure 5.13 shows this impact that this assumption has on the pipe wall temperature during the venting process. In cases with the soil layer included, the pipe walls temperatures were higher. This higher temperature was related to the thermal mass of the soil, which also cooled during the venting process.



Rev 01

Heavy Oil

07-2-LA-5507-0003

Figure 5.11 Schematic of heat transfer assumptions used in Olga models

Table 5.2 Material physical properties used in Olga modeling

Pipe Wall

Soil

Thermal Conductivity [W/m-K] 43 2.6

Density [kg/m3] 7,820 2,240

Heat Capacity [J/kg-K] 473 1,256



Rev 01

Heavy Oil

07-2-LA-5507-0003

Figure 5.12 Effective heat transfer coefficient for the different soil descriptions at steady state



Rev 01

Heavy Oil

07-2-LA-5507-0003

Figure 5.13 Prediction of pipe wall temperature for different heat transfer assumptions

5.2.2 Simplified flowline model for screening

Due to the detailed nature of the venting simulations, a series of simplified models were used in order to screen the most important parameters to be captured in the more detailed models that included the full flowline length and detailed topography.

Figure 5.14 shows the results of simulations for a series of different initial starting pressures. In some case the pressure was set to 39 bar, which was just above the single phase pressure and in other cases the pressure was set to 140 bar, which was defined as the maximum system operating pressure. In all cases, upon initiation of the venting process, the pressure rapidly dropped from the starting condition to the two-phase pressure defined by the CO2 phase

boundary. This suggests that the starting pressure was not important, thus all simulations were completed using an initial condition of 110 bar as per the design basis.

Figure 5.15 shows similar information to that in Figure 5.14, by showing the pressure- temperature during the venting process. The pressure decreased with minimal change in temperature until the system pressure reached the two-phase equilibrium line. After which the pressure and temperature decrease along this two-phase line until all the liquid was vaporized.



Rev 01

Heavy Oil

07-2-LA-5507-0003

This condition depends on the specifics of the venting process, but was primarily related to the rate of venting (i.e. vent size) relative the rate of heat transfer in the system.

Figure 5.14 Pipe wall temperature given different initial conditions



Rev 01

Heavy Oil

07-2-LA-5507-0003

Figure 5.15 Temperatures during venting and relation to the CO2 phase boundary

Figure 5.16 shows the impact that the vent size has on the pipe wall temperature during the venting process. With large vent sizes, the system pressure was decreased more rapidly. This decrease in pressure was due to the vaporization of the liquid CO2 and subsequent loss from

the flowline. This meant that the rate of venting was equivalent to the rate at which the liquid CO2 was vaporized. In order for the CO2 to vaporize, heat must be supplied from the

surroundings. At faster venting rates, less time was available to transfer heat from the surrounds, so the temperatures decreased to colder values to supply the necessary heat. At slower rates, heat can be transferred from further away from the pipe wall, so the soil temperatures stay relatively warmer.

The downside to the smaller vent size was that it takes longer to depressurize the system. In Figure 5.16, the minimum temperature roughly coincided with the time to vaporize all the liquid CO2 in the flowline. Smaller vent sizes result in higher minimum temperatures, but much longer

times to depressurize. In the system, a balance is required to determine an optimal balance between pipe wall temperature and venting time. Figure 5.17 shows the summary results of the minimum pipe wall temperature versus the vent size. Based on this simplified analysis, in order



Rev 01

Heavy Oil

07-2-LA-5507-0003

to stay above a minimum pipe wall temperature of -45°C, the vent size needed to be about 5”. These simulations served as the starting point for determining the required vent size during the detailed simulations.

Figure 5.16 Impact of leak size on transient pipe wall temperature



Rev 01

Heavy Oil

07-2-LA-5507-0003

Figure 5.17 Impact of leak size on minimum pipe wall temperature

Figure 5.18 shows an additional sensitivity done with the flowline length. Note that the minimum temperature was not too different for these cases. The difference in behavior for the longest flowline may be related to the total time for the system to depressurize. These longer times allowed more of the surrounding soil to cool which implies that the soil properties become more important in capturing an accurate representation of the system temperatures during the venting process.

This is in contrast with Figure 5.16, which showed the coldest wall temperatures for the faster venting process. The difference between these two examples is the rate of venting. With a smaller vent size, the surrounding soil can transfer heat at a rate sufficient to keep up with vaporization of the CO2. But for a given vent rate, the longer it takes to remove the liquid CO2

from the pipe, the cooler the liquid becomes.



Rev 01

Heavy Oil

07-2-LA-5507-0003

Figure 5.18 Impact of flowline length on pipe wall temperature

Based on the results from the simplified model, the most important parameters to correctly account for during the venting process were determined. The minimum temperature in the pipeline during the venting process was determined by the rate of CO2 vaporization and the rate

of heat transfer to the system. The vent size determined how quickly the CO2 was vented. An

accurate representation of the soil layers and properties was required to correctly model the system thermal performance accurately.

5.2.3 Detailed flowline model for venting

Based on the results from the simplified model, the detailed model was used to determine the specifics related to the venting process.

The detailed model was compared with the simple model for a comparable scenario (Figure 5.19). A similar trend was observed in that the rate of initial temperature decrease was similar. The simple model (also the shorter flowline) achieved a minimum temperature earlier and then the temperature began to increase. The longer flowline (80 km + well branches) took a longer time to depressurize the system and hence reached a minimum temperature at a later time. Similar to the simple model, the longer flowline resulted in a lower minimum temperature.

Also, the impact of vent size is shown in Figure 5.20 for a 4” and a 6” vent. As with the simple model, the larger vent size resulted in lower temperature and faster depressurization times. Also similar to the simple model, the vent size, in order to stay above a minimum pipe wall temperature of -45°C, was between 4” and 6”. Due to uncertainties in the model and soil



Rev 01

Heavy Oil

07-2-LA-5507-0003

properties, it is recommended to use a vent size of 4” or less to ensure that the temperatures in the flowline remain sufficiently high during the venting process.

Figure 5.19 Comparison between simplified and detailed model for a 4” vent



Rev 01

Heavy Oil

07-2-LA-5507-0003

Figure 5.20 Comparison between a 4” and 6” vent for the detailed model

Figure 5.21 Flowline temperature profiles during venting process (Scotford to Well 1)



Rev 01

Heavy Oil

07-2-LA-5507-0003

Figure 5.22 Vent rate for a 4” vent

Figure 5.23 Vent rate for a 6” vent



Rev 01

Heavy Oil

07-2-LA-5507-0003

The minimum pipe wall temperature was sensitive to the heat transfer assumptions. The ambient temperature (i.e. the temperature surrounding the soil layer), was defined as either - 40°C (ambient air temperature in winter) or 0°C (ambient soil temperature at burial depth during winter). In Olga, the initial temperature of the fluid was specified. The Olga model assumed a linear temperature gradient between the fluid and the ambient condition. It was not possible to describe the actual temperature gradient in the system in the soil layer, but the 0°C representation was probably the most realistic. If -40°C was used as the ambient temperature, the soil temperatures were unrealistically cold and allowed more heat to be transferred away from the pipe wall. Figure 5.24 shows that this assumption of the ambient temperature resulted in about an 8°C difference in the minimum temperature observed. Similarly, the assumption of a lower thermal conductivity resulted in a similar decrease in the minimum temperature.

Figure 5.24 Impact of soil properties and ambient temperature on pipe wall temperature

Figure 5.25 shows the vent rate observed for a 4”and 6” vent. In these cases, the rate of venting was largely similar. The initial rate was impacted by the vent size and time to open vent. But after the initial surge, the vent rates were very comparable. In these cases the rates were defined by the rate of heat transfer.



Rev 01

Heavy Oil

07-2-LA-5507-0003

Figure 5.25 Vent rate for 4” and 6”

One of the flow assurance aspects to be considered during venting is the temperature and pressure in the system relative the hydrate formation envelope. Upon venting, the temperature in the flowline dropped sufficiently low that hydrate formation was possible. However, the flow assurance risk associated with this step was small. The total water content of the CO2 was low

(<6 lbs/MMscf) which meant that the total volume of hydrate formed was small and highly unlikely to form in sufficient volume to block the flowline. Once the pressure of the flowline dropped sufficiently low, hydrates were no longer stable. And any hydrates that had formed during the venting process would melt.



Rev 01

Heavy Oil

07-2-LA-5507-0003

Figure 5.26 Temperature/Pressure conditions during venting and relation to phase envelope

5.2.4 Venting of Flowline Sections between LBVs

Several additional simulations were completed to look at the impact of venting only a single flowline section between the line break valves. The section between LBV-3 and LBV-4 was chosen because that represented the section with the largest potential to trap liquids. During the venting process, the liquid CO2 must vaporize along the phase boundary and as such, the

coldest temperatures are expected to be at the vapor-liquid interface, such as would be created in an area of high liquid holdup.

The coldest temperatures were observed at the end furthest from the vent line, which in this case was near LBV-4. Figure 5.27 shows the flowline profile as well as the temperature profile when the minimum temperature was observed. Similarly, Figure 5.28 shows the liquid holdup at the time when the minimum temperature occurs. When the system was vented from only one end (LBV-3 end), the valley closest to the LBV-4 end trapped a significant amount of liquid and as a result, that section was predicted to be very cold during the venting process. It appeared that there was sufficient sweep velocity in the other parts of the flowline to prevent significant liquid from accumulating and hence the low temperatures were not observed.

If the flowline was vented from both ends, the extremely low temperatures could be avoided. In this case, the flowline was vented from the LBV-3 end until prior to the minimum temperature being observed. Because the system minimum temperature followed the phase envelope, the system could be depressurized to about 10 bar, which corresponded to a vapor-liquid equilibrium temperature of about -40°C. Once a pressure of 10 bar was achieved, then the



Rev 01

Heavy Oil

07-2-LA-5507-0003

LBV-3 vent was closed and the remaining pressure was relieved from the LBV-4 end. This served to remove the localized pockets of liquid holdup from the system and prevent the low temperatures. Figure 5.28 shows that the liquid holdup was significantly reduced when following this procedure and as a result, the minimum temperature observed in the flowline was increased over the case of venting from only one end.

When venting the entire flowline, i.e. from Scotford to Well 5, similar behavior was not observed. The Well5 end was at a relatively high spot, so less liquid can be accumulated there. However, this assumed a relatively coarse geometry near the Well5 end. If even a few short valley sections accumulate liquid CO2, there exists the potential to achieve very cold temperatures.

What these simulations showed was that the in general, the venting process maintained temperatures above the material limit of the pipe except in localized spots, near the end of the flowline furthest from the vent. However, these low temperatures can be mitigated if venting procedure is followed then allowed for venting from both ends. The correct sequence and timing of the venting at each end will need to be addressed once a more detailed plan for how the system will be vented is determined.

Figure 5.27 Minimum temperatures during venting of section between LBV3 and LBV4



Rev 01

Heavy Oil

07-2-LA-5507-0003

Figure 5.28 Holdup at minimum temperature during venting between LBV3 and LBV4

5.3 Wellbore Venting and Placement of Subsurface Safety Valve

A series of simulations were completed looking at the wellbore venting process in more detail in order to determine the placement of a subsurface safety valve. Initial work (1) on the venting process during a well blowout focused on the steady state temperature and liquid levels in the well. This work expanded that to include the transient effects associated with a well blowout. In these cases, it was assumed that the wellhead was removed from the well, so there was a full well diameter leak path from the reservoir to the atmosphere.

The placement of the subsurface valve depended on the timing in which it was closed. In this work, two criteria were investigated. The first criteria was based on placing the valve in the single liquid/dense phase region so that upon valve closing, there was not a significant pressure drop across the valve. The second criteria was based on the hydrate stability region. The



Rev 01

Heavy Oil

07-2-LA-5507-0003

desire would be to place the valve in a region where hydrates are not stable because the presence of hydrates may prevent the valve from closing fully.

In these simulations, fluid was injected into the well, to establish steady state injection condition, at the full design rate of 1.2 Mtpa, which established the coldest well temperatures. At time zero, the well all flow was stopped. And at minute 6, a blowout of the well was started. The 6 minute lag between shut-in and blowout were to aid in the simulation stability. The blowout was initiated to go from fully closed to fully open in 5 seconds. Figure 5.29 and Figure 5.30 show the liquid level in the well as a function of time. Initially the liquid level was about 150 m into the well. During the blowout the liquid level began to drop and over the course of about 5 minutes, the liquid level dropped by about 500 m. If the time scale is expanded, Figure 5.30, the liquid level continued to drop for several hours before reaching a final depth of about 1100 m.

Similarly, temperature isotherms can be plotted for the well, as seen in Figure 5.31 and Figure 5.32. Note that temperatures in the well began to recover after some time. Initially there was a considerable amount of JT cooling occurring in the well due to the rapid depressurization. It took some time for the relatively warmer reservoir fluids to warm the system. As seen in Figure 5.32, the coldest temperatures occurred early in the blowout process. In this figure the two most relevant temperatures are based on the hydrate equilibrium temperature for a fluid with 4 lbs/MMscf (0°C) and with free water (10°C). During the initial blowout, the 4 lbs/MMscf defined the relavent hydrate equilibrium condition. As the blowout continued, it was expected that aquifer water would be produced along with the CO2, in which case the free water

temperature was more relevant. It was unknown how long it would take to see free water during a blowout, but it is likely to greater than 5 minutes.

Figure 5.33 shows a summary chart of the valve setting for the different criteria. The liquid level criteria defined the deepest setting requirement for the valve. It was previously assumed that it would take some time to begin producing free water. If the time to produce free water was greater than about 5 minutes, then the valve setting required is always determined by the liquid level in the well. It should also be noted that the fluid surge impact on the bottomhole pressure also needs to be taken into account.



Rev 01

Heavy Oil

07-2-LA-5507-0003

Figure 5.29 Liquid level in well during blowdown (initial time)

Figure 5.30 Liquid level in well during blowdown (steady state)



Rev 01

Heavy Oil

07-2-LA-5507-0003

Figure 5.31 Temperature isotherms in well during blowdown (steady state)

Figure 5.32 Temperature isotherms in well during blowdown (initial time)



Rev 01

Heavy Oil

07-2-LA-5507-0003

Figure 5.33 Safety valve setting based on single phase and hydrate criteria

Table 5.3 Summary of valve closing time base on single phase criteria in well

5.3.1 Blowout in a well with a low reservoir productivity



Rev 01

Heavy Oil

07-2-LA-5507-0003

As a sensitivity, the impact of a lower reservoir productivity was investigated. When the reservoir productivity is lower, the bottomhole pressure is lower, which means the average pressure in the well is lower. This lower pressure results in a lower liquid level as well. Figure 5.34 shows a comparison between the base and low PI cases. For the first minute of the blowout, the depth setting requirement is about the same. For longer times, the lower PI case requires the valve to be set deeper. Depending on the reservoir characteristics and the valve closing time, this may be an issue and should be incorporated into the final design of the subsurface safety valve.

Figure 5.34 Impact of reservoir injectivity on safety valve setting



Rev 01

Heavy Oil

07-2-LA-5507-0003

6.0 REFERENCES

1. Dykhno, Leonid, et al. Quest CCS Prospect: Flow Assurance for System Selection. s.l. : Shell, 2010. GS.10.53258.

2. Peters, David, et al. Quest CCS: Flow Assurance - ITR 13-17 June 2011. [powerpoint presentation] 2011.

3. Hugonet, Vincent and Perez, Carlos. Assumptions for system transient analysi Update - 6 Apr 2011. [email/document] 2011.

4. Hugonet, Vincent. Geothermal gradient. [Excel spreadsheet] 2010.

5. Song, Kyoo and Kobayashi, Riki. RR-99: The water content of CO2-rish fluids in equilibrium with liquid water and/or hydrates. 1986. GPA Project 775-85.

6. Hydraulic Guidelines. s.l. : Shell, 2009. GS.09.51858.

7. Peters, David, Lacy, Rusty and Dykhno, Leonid. Quest CCS Prosect: Flow Assurance Evaluation of Low Flow Events. 2011. intermediate report.



Rev 01

Heavy Oil

Quest Pipelines Flow and Flow Assurance Design …Quest Pipelines Flow and Flow Assurance Design and...

Documents

Transcript of Quest Pipelines Flow and Flow Assurance Design …Quest Pipelines Flow and Flow Assurance Design and...