Offshore Technology Conference issue

—
Baldor Electric Company is now ABB
A history of innovation. A future of opportunity.

For more than seven years, Baldor Electric Company has been a member of
the ABB Group.

But as of March 1, 2018, we ARE ABB, and we are proud to go to market now
as one company.

What does that mean for our customers?

- The same Dodge® mechanical power transmission
products and Baldor-Reliance® motors you trust and prefer
- The quality and dependability you deserve
- Local sales, support and services
- Local manufacturing and inventory
- Industry expertise
- Reliable solutions
- The look and feel of ABB
- Doing business with one Forbes Global 250 company
- The global support and technology of ABB
- A future of opportunity and innovation

Thank you for your business. We look forward to continuing and growing
our partnerships for many years to come.

Let’s write the future. Together.

Baldor.com/WeAreABB
 Get accurate flare
gas heat values fast,
with the FlarePro™ Mass Spectrometer

With impending EPA rule changes for flare gas stack combustion, you’ll need to measure
heat value at the stack quickly, accurately, and often. AMETEK Dycor®’s FlarePro mass
spectrometer has the speed and specificity for the task, calculating make-up gas quantities
10X faster than chromatographs. It can speciate and quantify alkanes and alkenes up to
C7 and convert ASTM calculations of concentration to yield flare gas BTU values, even with
wildly changing flare gas streams. Features include:

• 24” footprint (general or C1, D2 areas) • Low maintenance

• 16-valve inlet system
• 1 to 100 amu mass range
• Easy-to-service modular design
Learn more at www.ametekpi.com.
• Modbus/OPC communications
• Environmentally controlled to
accommodate a wide range
of temperatures

© 2018 by AMETEK Inc. All rights reserved.

All your 3R nee

recision cut
ring p
offe
Now
ds in one place.

segments 12
” thro
ugh
48”

weldbend.com
3010 HIGHLAND PARKWAY, STE. 325
DOWNERS GROVE, IL 60515
630-571-4070, FAX 630-214-4504
CONTENT SPECIALISTS/EDITORIAL
KEVIN PARKER, Senior Contributing Editor
630-890-9682, KParker@CFEMedia.com
EMILY GUENTHER, Associate Content Manager
EGuenther@CFEMedia.com
KATIE SPAIN, Art Director
KSpain@CFEMedia.com
PUBLICATION SERVICES
JIM LANGHENRY, Co-Founder & Publisher
JLanghenry@CFEMedia.com
STEVE ROURKE, Co-Founder
SRourke@CFEMedia.com
TRUDY KELLY, Executive Assistant
TKelly@CFEMedia.com
AMANDA PELLICCIONE, Director of Research
APelliccione@CFEMedia.com
ELENA MOELLER-YOUNGER, Marketing Manager
EMYounger@CFEMedia.com
KRISTEN NIMMO, Marketing Manager
KNimmo@CFEMedia.com
PAUL BROUCH, Director of Operations
PBrouch@CFEMedia.com
CHRIS VAVRA, Production Editor
CVavra@CFEMedia.com
MICHAEL ROTZ, Print Production Manager
717-766-0211, Fax: 717-506-7238
mike.rotz@frycomm.com
MARIA BARTELL, Account Director, Infogroup Targeting Solutions
847-378-2275, maria.bartell@infogroup.com
RICK ELLIS, Oil & Gas Engineering Project Manager,
Audience Management Director
303-246-1250, REllis@CFEMedia.com
LETTERS TO THE EDITOR
Please e-mail your opinions to KParker@CFEMedia.com
INFORMATION
For a Media Kit or Editorial Calendar,
e-mail Trudy Kelly at TKelly@CFEMedia.com
REPRINTS
For custom reprints or electronic usage, contact:
Marcia Brewer, Wright’s Media
281-419-5725, MBrewer@wrightsmedia.com
MAILING ADDRESS CHANGES
Please e-mail your changes to customerservice@CFEMedia.com
PUBLICATION SALES
JUDY PINSEL, National Sales JPinsel@CFEMedia.com
3010 Highland Parkway, Ste. 325
Downers Grove, IL 60515
4 • APRIL 2018 OIL&GAS ENGINEERING
AI on a chip; AI in the cloud
I
n its efforts to build next-generation
remote robotics controls, TechnipFMC
has selected Real-Time Innovations’ RTI
Connext DDS as the connectivity frame-
work for its underwater remotely operated
vehicles (ROVs).
Real-Time Innovations (RTI) says it’s the
industrial Internet of Things (IIoT) “connec-
tivity company” and the largest supplier of
products based on the Object Management
Group’s data distribution service (DDS) stan-
dard. In 2017, RTI introduced RTI Connext
DDS 5.3, said to be the first connectivity
software for scalable system-of-systems
database architectures.
The news of the moment is that RTI
will supply Connext DDS as an appropriate
engine for TechnipFMC’s control system,
but what’s behind it is the proliferation of
artificial intelligence (AI). In fact, RTI says
it “lives at the intersection of functional AI
and pervasive networking.”
A kind of bus
This purely software connectivity engine is
a kind of bus that sits “above” the physical
network. It is used by software developers,
including equipment builders and systems
integrators, for uses associated with
drones, autonomous vehicles, medical
devices, and other systems requiring low
latency combined with high reliability and
resilience, said David Barnett, vice president
of products and markets, RTI. Latency is in
microseconds, allowing millions of messages
per second.
847-624-8418 While Big-Data applications are likely cloud-
Fax 630-214-4504
bound, real-time applications are more likely
found on a chip, although not necessarily so.
“The architecture is completely distributed
and peer-to-peer. A more typical bus would
have a message broker that forwards the data
to the controller,” Barnett said. “Here a sen-
sor and controller exchange the data directly.
There is no single point of failure.”
RTI worked with the Object Management
Group to define the DDS standard and now
adheres to it. “Besides base connectivity,
COMMENT
KEVIN PARKER
SENIOR CONTRIBUTING
EDITOR
we also provide the
infrastructure that
surrounds it and
that ensures per-
formance, including
debug and test, scalability, and enhance-
ments,” Barnett said. For example, what’s
required to manage hundreds of thousands
of medical devices is already in the solution.
“The big opportunity is the move to
autonomous systems,” said Barnett. “The
most common IIoT applications tend to take
data from the edge, put it in the cloud, and
apply it to something like predictive mainte-
nance. Autonomous applications don’t want
to go to the cloud. We’re talking about highly
distributed control systems, which require
distributed intelligence.”
In the skies, under the seas
Barnett noted how airborne drones and
underwater ROVs both offer similar
challenges. “What RTI provides falls outside
the typical enterprise. The DDS protocol
allows real-time quality of service tuned for
a wider range of networking environments.
In the subsea, for example, packet delays
can be sent so that instantaneous response
isn’t needed. The application itself can define
real-world constraints. Finally, if there is a
network problem, the application can be put
in a safe, secure mode, as needed.”
Its developers say DDS is the only frame-
work that provides a peer-to-peer, publish/
subscribe architecture. With DDS, each
application participates in a data bus that
comprises a set of data topics, each defined
with its unique, discoverable data model.
The Object Management Group manages
both the Industrial Internet Consortium and
its standards area. The RTI website notes
that the OPC Foundation has come a long
way in its understanding of the pros and
cons of DDS and OPC-UA. OG
I NSIDE Cover photo courtesy Maersk. Deep Sea Mooring, a Vryhof company, has
commenced operations in providing mooring equipment, design, and marine
supervisory services to Maersk Drilling in Trinidad. Shown is the deepwater semi-
submersible rig Maersk Developer, which covers drilling activities in block E of
the East Coast Marine Area offshore Trinidad.

6
FEATURES
6 Wellsite valve manifolds simplified
A multiport flow selector is safer, less expensive, lighter
weight, and more compact then conventional units

10 Recent advances in shale field

flowmeter proving
Mass and density calibration can now be done in the field

14 How machine learning contributes

to smarter pipeline maintenance
Pipeline integrity management is an industry
challenge, but a technology breakthrough can help

20 Improve the accuracy

of underwater cathodic 10
protection surveys
AUVs and improved analytics offer effective
method for pipeline inspection

25 Some alternatives for improving

an octane upgrade process
A properly optimized unit can recover 95%
of normal paraffins in the feed

20

OIL&GAS ENGINEERING APRIL 2018 • 5

FLOW CONTROL AND MEASUREMENT
Wellsite valve manifolds simplified
A multiport flow selector is safer, less expensive, lighter weight, and more compact
than conventional units
C
By Joseph Zawacki onnecting multiple production
wells or flowlines to a test sepa-
rator normally requires multiple
valves, each of which must be
opened by hand to connect each
well in turn. In addition to valves
and shuto-ff devices, such a system requires
considerable piping (see Figure 1).
Conventional systems cost more initially,
entail higher operating costs, lead to mainte-
nance and downtime, and add to personnel
risks. A multiport flow selector (MPFS)—with
automatically actuated control—is safer, less
expensive, lighter weight, and more compact
than conventional systems.
Let’s look at how a MPFS works and detail
its advantages over conventional installations.
Well testing
Testing constituent flows is a crucial part
of an oil production
process. Oil, water,
brine, condensate,
gas, and other fluids
must be monitored and
measured regularly. In
addition to separating
the well stream
into its constituent
components, the well
test separator records
the volume of each
component over time,
allowing the flow
rates to be calculated,
most often as barrels
per day (b/d) for liquid
products or million
standard cubic feet
Figure 1: Piping from well
sites to a test separator is
shown. All images courtesy:
Fisher Controls
6 • APRIL 2018 OIL&GAS ENGINEERING
per day (MMSCFD) for gas products. These
calculations allow the tester to determine
how productive a well is. They furnish insight
into the underground reservoir of oil or natural
gas into which the well is tapped.
End users select how sequencing is man-
aged, either on-demand or scheduled. For
scheduling, a separate controller executes the
logic that controls actuator operation. A test
separator requires settling time between well
switching, so a typical system would expect
well-test frequency of about one well per day.
Historically, wells are connected to a com-
plex piping network with two manifolds. One
manifold connects all wells to production, and
the second connects the well selected for
testing to a shared multi-phase flowmeter or
separator. While this conventional approach
uses common valves, actuators, and piping
components, it is more expensive and electri-
cally complex than a multiport system.
For example, a conventional seven-well,
two-way well-test manifold requires 21 valves
(Figure 2). During normal operation, all the
wells are connected to common production
and test lines. If Well 1 needs to be tested,
a technician manually closes the production
valve, then manually opens the valve to the
test separator. Well 1 is then directed to the
test facility, while all other wells continue
feeding the production line.
Depending on the test equipment, flow
set up, and separator size, a test can take
several hours. Size needs to be optimized
for the wells’ production rates. Issues can
arise if the test separator is undersized.
For one, time to execute a change-over is
longer for an oversized separator. A rule of
thumb is that it takes 24 hours for the the
separator contents to turn over.
While tests can take less time, the oper-
ators most likely are on a rotational visit
cycle for the field. The change-over takes
too long for them to sit and wait, so test
time usually is dependent on the field visi-
tation rotation.
Typically, operators reset a flowmeter on the
downstream side of the separator (on the oil
outlet). They either assume a certain amount
of flow—perhaps a time-based calculation
based on previous wells’ flow rates—or they
come back after the separator has stabilized
with the new well, reset the flowmeter, and
officially start the test.
All of this takes quite a bit of and time, and
as a result, owner-operators don’t get very
frequent test data for wells with conventional
manifold systems.

Automating tests
The wells of a typical onshore oil & gas and managing those wells to maximize pro- Figure 2: A standard well
production site are scattered over a large area, duction throughout that lifespan is a critical manifold diverts oil to pro-
so it is difficult to manually open and close the activity for profit/cost management. duction via the green valves,
and to a test separator via
manifold on-and-off valves. Oil fields are found
the red valves. For a seven-
in remote deserts, deep jungles, or in sub-zero Multiport flow selectors
well system, this arrange-
environments. These conditions can make A typical MPFS has eight inlets and two
ment requires 21 valves.
it difficult for technicians to access valves. outlets (Figure 4). Of the eight inlets, seven
Offshore platforms often are unmanned. In connect to the wells, while the eighth
those cases, maintenance requires a boat connection is generally used as a parking
or helicopter trip. Therefore, well operators location or an observation port for the
automate test-flow diversion. selector plug. This allows for temporary
An MPFS that’s controlled remotely saves maintenance and flushing. It allows
money by reducing the number of valves and production from all seven wells if the test
the labor to operate the valves, as well as separator system is offline. One outlet
improving safety. For example, a seven-well connects to the test system, and the other
Figure 3: By using a multi-
system with a MPFS requires only seven outlet carries the flow of all other wells to
port flow selector, diverting
valves (Figure 3) as compared to 21 with a the production header. flow to the test separator is
conventional system. It eliminates the need Normally the pressure drop across an automated and the number
for operators to open and close valves manu- MPFS is very low because the flow streams of valves needed is reduced
ally, and it reduces valve maintenance. are only diverted, not restricted. In some to seven.
Even with an MFPS, tests are done about
once per day, typically on a rotating schedule
from well to well. However, an operator can
initiate a test based on anomalous test data,
in preparation for work on a well, immediately
after such work, during early well-life produc-
tion when monitoring water cut-back, and for
artificial-lift management.
If artificial-lift management is being applied
to a well, it needs to be monitored for correct
operation and to maximize reservoir produc-
tion. Continuous analysis may be needed, and
this often requires separation tests.
Fundamentally, more data is great for res-
ervoir engineers—especially in shale applica-
tions when well characteristics change rapidly.
Shale wells might only have a three-year life
OIL&GAS ENGINEERING APRIL 2018 • 7
FLOW CONTROL AND MEASUREMENT
Figure 4: Multiport flow
selector installed on an off-
shore platform.
Figure 6: Shown is the con-
trol panel of a Bettis multi-
port flow selector actuator.
8 • APRIL 2018 OIL&GAS ENGINEERING
applications, flow velocity will limit the
maximum flow or maximum number of
connected wells, but this condition is no dif-
ferent than when using a conventional dual
manifold system.
An electrical actuator with electronic con-
trols and limit switches enable easy control
and monitoring of an MPFS. Whenever well
testing is performed, operators simply issue
a remote command to position the diverter
plug to the desired well inlet. This reduces
the chances of manual error that may lead to
production downtime and incidents.
The commands can be sent from a central
control room to remote wells via wireless,
cellular, or direct-wired connections. The actu-
ator also can be controlled
locally. A Fisher multi-port
valve actuator (MPA) can
integrate with Emerson’s
DCMlink software for con-
figuration and diagnostics
activities and to a distrib-
uted control system (DCS).
The DCS hosts the opera-
tional software, sends com-
mands to the actuator, and
manages the schedule for
automated testing cycles.
An MPFS greatly simpli-
fies piping, especially on
Figure 5: The test separator on this offshore platform
services 14 wells by using two multiport flow selectors
mounted side by side.
cramped offshore platforms (Figure 5). It
also eliminates the need for a technician to
open and close valves in a dangerous envi-
ronment.
Operation can be done remotely, as an
MPFS actuator supports numerous com-
munication protocols, including Modbus,
Modbus TCP/IP, Profibus DP, Foundation
Fieldbus, DeviceNet, and Ethernet. The
MPFS controller provides outputs such as
alarms, torque, flow, and port selected.
Built-in diagnostics check for issues like con-
trol error, motor overload, and power loss.
The MPFS can be controlled locally from a
panel (Figure 6).
An MPFS provides an additional advantage
in sour oil & gas fields, which often require
special materials at points of contact. For
example, when high nickel-chromium inter-
nal cladding is required, the simplified piping
and fewer valves result in much lower clad-
ding cost. OG
Joseph Zawacki is vice president of the strategic
business unit, hydraulics at Emerson, where he
manages the high-pressure and hydraulic-product
portfolio team. He holds a MSME and an MBA from
the University of Michigan.
 We understand how you need to reduce
complexities at your plant.

CLEAN PROCESS
+ CLEAR PROGRESS You strengthen your plant’s safety, productivity,
and availability with innovations and resources.

Promass Q – for increased plant productivity

• Error-free flow measurement in custody transfer applications

in mass or volume units due to unmatched accuracy for density
determination
• Ideal for hydrocarbons with entrained gas thanks to the
patented Multi Frequency Technology (MFT)
• Patented “Heartbeat Technology” for device verification during
operation and permanent self-diagnostics

Visit us during OTC 2018

Booth 105

Do you want to learn more?

www.us.endress.com/promass-q300
UPSTREAM INSTRUMENTATION
Recent advances in shale-field
flowmeter proving
Mass and density calibration can now be done in the field
By Brian Hoover
Figure 1: Calibrating flow-
meters in remote locations
like the Permian Basin can
be challenging, but solutions
are available in the form of
local instead of remote cali-
bration. All images/graphics
courtesy: Endress+Hauser
10 • APRIL 2018 OIL&GAS ENGINEERING
A
revival in U.S. oil & gas production
is driving increasing numbers of
allocation and custody-transfer meter
installations. These installations, in
turn, mean an increased need for
“proving” or calibrating flow-mea-
surement points used for transactional purposes.
Meeting these calibration requirements in oil &
gas fields (Figure 1) is particularly difficult because
shale plays are often in isolated, remote locations,
far from proper maintenance facilities.
Estimates say more than 5,000 wells will be
drilled in the Permian Basin over the next 20
years. Each well will have one or more flowme-
ters that need to be calibrated and proven.
Flowmeter proving must be conducted on
a regular basis based on regulations and legal
contracts. The point is to validate accurate inter-
nal product transfer. Meeting Environmental
Protection Agency (EPA) regulatory requirements
for reclaimed water used in fracking also entails
frequent flowmeter testing.
This increased demand, coupled with a short-
age of staff and systems, has created a “calibra-
tion crisis” in many shale fields. The solution to
this problem lies with better calibration methods.
Rather than removing flowmeters for ship-
ment to calibration labs, recent developments,
now being deployed in the Permian Basin, bring
flowmeter calibration to the shale fields.
Calibration requirements
Shale field flowmeters are calibrated, typically
based on the contract between the operator
and its customers, about every three months.
The Bureau of Land Management (BLM)
recently released regulations specifying
calibration frequency based on production-
well characteristics. BLM specifies calibration
frequency, and monthly or quarterly proving
intervals are typical. Custody-transfer allocation
locations are the most critical, while loading rack
or regulated meters are often calibrated annually.
Flowmeters are proven to identify any signifi-
cant shift in meter factors that result in greater
measurement uncertainty. When a shift occurs, a
root-cause analysis usually is conducted. Changes
can be due to paraffin buildup, mechanical dam-
age to the fiscal meter, or upset operating condi-
tions, such as gas carry-under in liquid lines.
Typically, calibrations are done on a regular,
scheduled basis, unless there is a dispute or rec-
onciliation problem. Unless flowmeter removal
requires shutting down a wellsite or facility,
meters are proven during operations.
When a flowmeter has to be sent to a calibra-
tion lab, the meter is removed, cleaned, packed
up, shipped off to the lab, returned, and re-
installed. This process can take a week or more
for each flowmeter, at a cost of about $500 per
meter just for calibration—with additional costs
for shipping, time, and labor. In some cases, the
wellsite or facility is shut down pending return of
the calibrated meter.
More often, critical meters are calibrated or
proven “in-situ” without being removed from
service. Most installations have proving taps to
facilitate this process by allowing quick connec-
tion to a field prover, while others do not and
require removal of the meter from the process.
Field proving and calibration typically are done
through third-party service providers that travel
to well sites to do the work. Large end users
may have a prover integrated into their metering
systems. Alternatively, field references such as
ball provers (Figure 2), small volume provers, and
master meters can be used.

Doing it in the field

Field provers compare a known volume against
the meter output. The ratio between the prover
reference volume and the meter reading is the
meter factor, used to correct the meter reading. Instead of relying on ball provers, piston prov- Figure 2: A ball prover works
Provers can be uni-directional or bi-directional ers, or master meters, recent developments by passing a known volume
and use a sphere (ball provers) or a piston make it possible to bring a “calibration lab” to of gas or oil through a meter
(piston provers). A flowmeter also can be put in the shale field. For example, Endress+Hauser is while monitoring the meter’s
series with a master meter, with their readings working with a third party to deploy their HP80 output. The yellow ball (top)
is a sphere containing the
compared. (See Figure 3.) (U.S. patent pending 15/605, 562) Field Reference
known volume.
Field references used for proving typically have Meter Standard (Figure 4) for mass, density, and
a higher level of uncertainty when compared to volume determination under existing operating
stationary calibration facilities or provers, are prone conditions. This solution also is being adopted for
to site constraints, and don’t work well in less calibration of tanks and level instruments.
than ideal operating conditions. Conventional ball
provers and compact provers have many mechani- Field reference meter standards
cal components—such as a four-way valve, piston This type of field reference meter standard
seals, motors, and pumps—that must be main- system provides field metrologists and
tained and are subject to mechanical wear. calibration specialists with the metrics, tools,
Master meter proving normally has the high- and information needed to measure and manage
est total uncertainty of all meter proving meth- all primary measurements in the upstream,
ods. The technique used to prove the master midstream, and downstream segments of the oil
meter and the process to prove the line meter & gas market, including shale fields.
introduce various levels of uncertainty into the In addition to measuring the accuracy of the
petroleum measurement hierarchy. Some of the mass flow, operating density, volume flow, linear-
factors that can contribute to a higher uncer- ity, and repeatability of an equipment under test
tainty include the following: (EUT), a field reference meter standard also can Figure 3: Like a ball prover,
• Installation conditions where the master capture and report field conditions including infor- a piston prover provides a
meter is not proven in-situ mation on the process temperature, pressure, known volume of oil or gas
• Differences between the viscosity and den- Reynolds number, and viscosity. for calibrating a flowmeter.
sity of the liquid used to prove the master
meter and the liquid used during proving
• Differences between the temperature, pres-
sure, flow conditions, and flow rates used to
prove the master meter and those present
during line meter proving
• The reproducibility of the interval between
proving, severity of service, meter damage,
transportation and storage, use, corrosion,
and other functions
• Flow rate changes during proving of the mas-
ter meter that result in poor repeatability or
bias errors due to delay in response time of
the master meter pulse output. Larger prover
volumes may reduce the effect because it
increases the proving time.

OIL&GAS ENGINEERING APRIL 2018 • 11

UPSTREAM INSTRUMENTATION
Figure 4: A field reference
meter standard mounted on
a trailer brings the capabili-
ties of a calibration lab to the
shale field.
Figure 5: Hydrocarbons pass-
ing through the equipment-
under-test (EUT) flowmeter
(at the bottom) is routed
to field reference Coriolis
flowmeters (center) for veri-
fication.
12 • APRIL 2018 OIL&GAS ENGINEERING
These systems provide in-situ verification and
calibration of all types of flowmeters including
ultrasonic, positive displacement, turbine, and
Coriolis flowmeters.
For example, the HP80 system can calibrate
1.5 in. to 4 in. nominal size flowmeters over a
range of 35 to >750 gallons/minute. The system
uses three 2-in. Endress+Hauser Promass Q
Coriolis flowmeters as field reference standards.
The system control valves allow the user to con-
figure one, two, or three meters for the test set
up (Figure 5).
To conduct a flow verification, the team posi-
tions the field reference meter standard system
as close as possible to the piping containing the
EUTs. A field technician connects hoses using
swivel joints to the EUT’s prover taps, opens the
control valves, and adjusts the flow to accommo-
date calibration of the EUT.
Hydrocarbons flow through the EUT, and then
through the Coriolis field reference meters. A
flow computer accumulates the totals from the
reference meters in the system and compares
the EUT output to that of the reference standard.
A meter factor and repeatability calculation are
applied to complete a successful prove. The new
meter factor is compared to the previous meter
factor to assess any trends. A successful prove is
dependent on obtaining three to five consecutive
meter factors within 0.05%.
The calibration records are managed using
a flow computer, with process data recorded
and managed via the Endress+Hauser RSG45
advanced data manager, or similar software.
The system is self-contained, including a 24 Vdc
power supply for all field and reference devices.
Solar panels are used to maintain a charge within
the battery assembly. Calibration profiles and
results are captured and reported and can be
archived locally on secured media, communicated
to a local network for storage, or sent to cloud
storage as part of an Industrie 4.0 solution.
A typical calibration takes 30-45 minutes per
meter including connection, test report generation
and disconnection time. Field trials have shown
these types of field reference meter standard
systems can calibrate two to four times faster
than conventional ball provers or piston provers,
with much greater accuracy on a mass basis.
Full-stream density measurement is used to
ensure product quality and to facilitate volumetric
calibration using operating conditions or American
Petroleum Institute (API) tables.
Summary information
Expansion of shale oil & gas fields in the U.S.
have increased the need for regular calibration
of flowmeters and level devices. Conventional
calibration methods are expensive, time-
consuming, and with entail levels of uncertainty.
Recent developments now bring laboratory
quality calibrations to the shale fields with field
reference meter standards systems deployed
through third-party service providers. OG
Brian Hoover is the business development manag-
er for Strategic Alliances at Endress+Hauser, where
he is active in developing solutions for managing
gas, oil and, water.
Connecting What’s Needed with What’s Next™

Visit us at OTC
Booth 2060
MAKE INFORMED
DECISIONS, REDUCE
YOUR COSTS

Copyright © 2018 Oceaneering International, Inc. All rights reserved.

We do things differently, creatively, and smarter by providing subsea surveys and deepwater seafloor mapping
solutions. As your trusted subsea partner, our unmatched experience and breakthrough technologies enable us
to adapt and evolve regardless of market conditions.

By working together, we will safely and reliably re-shape the future of the oil and gas industry.

Connect with what’s next at oceaneering.com

AUTONOMOUS VEHICLES & SUBSEA ANALYTICS
Improve the accuracy of underwater
cathodic protection surveys
Autonomous vehicles and improved analytics offer effective methods for pipeline inspection
T
By Adam Haavisto o reduce risk and increase inspec-
tion speeds—while lowering
costs—for underwater pipeline
surveys, new inspection methods
must be considered. Current meth-
ods are slow, capital intensive, and
tend to produce a low value-to-cost ratio.
In fact, pipeline surveys rarely identify
issues, but are nevertheless required,
either through regulations or company
guidelines. Therefore, they are a sunk cost
that needs to be reduced, and autonomous
underwater vehicles (AUVs) may provide
the means.
Oceaneering is developing a method for
inspecting subsea pipelines that uses AUVs
combined with analytical techniques and
machine learning. The aim is to improve
data quality and reduce costs compared
14 • APRIL 2018 OIL&GAS ENGINEERING
to the current practice of using a remotely
operated vehicle (ROV) to perform pipeline
and cathodic-protection (CP) surveys.
The overall goal is to reduce offshore time,
which is the largest cost factor, while still
maintaining the data quality of ROV surveys.
Yesterday and today
Traditionally, long pipelines are inspected
with an ROV. The ROV is operated by a
pilot on the service vessel, an expensive
proposition in and of itself. The benefits of
this method are instant feedback, the ability
to reroute (or sidestep) for spans, inspection
flexibility, and direct CP measurements.
The challenges associated with this method
are pace, weather, and poor track records
for performing CP surveys, as well as ill-
equipped ROVs.
In this scenario, the ROV
sits on top of the pipeline in
order to walk it, taking remote
electrode measurements and
pipeline footage and data.
Besides extended use of
the service vessel, the ROV
relies on cameras to perform
visual inspections, along with
direct stabs that make physi-
cal contact with the pipeline
The C-Surveyor AUV fleet can provide
high-resolution pipeline inspection
services, equipped with a Kongsberg
Simrad EM2040 Multibeam Echo-
sounder, high-resolution digital cam-
era, 2G Robotics high-resolution laser
bathymetry system, magnetometer,
and a GeoChemical suite of sensors.
The geo-referenced still photo imagery,
laser data, and multibeam data allow
for change detection across surveys. All
photos courtesy: Oceaneering
to measure CP voltage. The process is time
consuming. The ROV collects limited data
and is hampered by the pipeline’s protective
coating.
Recently introduced ROV-based systems
are said to increase accuracy and speed.
However, results presented are often best-
case scenarios, leaving clients frustrated
with actual outcomes.
The alternative approach under dis-
cussion employs an AUV equipped with
lasers, high-definition imaging, and field-
gradient sensors. The AUV automati-
cally follows the pipeline, gathering
data from a distance above the sea-
bed of about 1 meter to 2 meters,
or 3 ft. to 7 ft. Analytical methods
(including visual processing, shape
filling, trending, and mechanistic
calculations) are used to evaluate
pipeline integrity and identify areas
that are prone to failure.
Automation of the survey increas-
es speed and minimizes personnel,
thereby reducing errors and costs.
Solving issues
By combining AUV inspections
with high-definition imaging
technologies, it is possible to take
pipeline scans and images and
place them into an algorithm that
will identify various aspects of the
pipeline.
As noted, with current meth-
ods this process is slow, because of
the hours involved and because some-
one must manually review the footage
and mark down features. Current anode
inspection involves a subjective grading
system, where the operator views the
anode and assigns a grade in 25% incre-
ments, which may not be accurate.
Alternatively, the incorporation of
machine learning into the AUV-driven pro-
cess means that, via the algorithm, the
machine is taught to do the processing
needed, identifying anodes, free spans,
and scars along the pipeline. “Teaching”
involves feeding the machine with hun-
dreds of hours of correctly identified
pipeline features, such as anodes, scars,
and free spans.
Additionally, a point cloud system deter-
mines anode degradation and pipeline con-
dition. With multiple surveys over several
years, it’s possible to see how the anode
degrades over time, as well as display more
quantitative data. For pipeline conditions,
scars, free spans, and lateral movements can
be tracked over the life of the pipeline, there-
by identifying for engineers those trends that
could lead to issues. Using point cloud data,
degradation can be quantified more accu-
rately, for greater survey confidence.
ROVs versus AUVs The AUV can automatically
Three distinct cost categories impact the two follow the pipeline, gather-
different survey mechanisms and methods. ing data from 1 to 2 meters
These are the preparation, execution, and above the seabed. Analytical
methods, including visual
reporting of the work.
processing, shape filling,
Preparation includes any work involving
trending, and mechanistic
mobilization of equipment and personnel.
calculations, are used to
Here there is minimal difference in cost evaluate pipeline integrity
between ROVs and AUVs. and identify areas that are
Execution of the work is where AUVs can prone to failure.
deliver the most savings compared to ROVs.
Using an AUV is a faster process because
AUVs move at higher speeds compared to
ROVs, without reducing the data quality. An
AUV can reduce offshore time by 40%.
Use of the AUV does entail an increase
in time needed for onshore analysis and
reporting. The anticipated increase for work
OIL&GAS ENGINEERING APRIL 2018 • 15
AUTONOMOUS VEHICLES & SUBSEA ANALYTICS

reporting of about 50% is due to increased

engineering hours to quality check the auto-
mated data. However, overall, AUV use is
still cheaper on a per-hour basis because the
largest cost component per project is the
boat rental time.
On average, Oceaneering estimates a
total survey savings of 30% to 40% using
the AUV and incorporating tools such as 3-D
imaging and high-definition algorithms. Thus,
AUV-based inspections can be performed in
one-third the time and at a lower cost com-
pared to ROV inspections.
In addition, regularly scheduled surveys
performed using AUVs enable engineers to
overlay multiple data sets to evaluate CP
anode degradation rates and pipeline integ-
rity over the pipeline’s planned lifetime. This

AUVs move at a higher speed than ROVs without reduc-

ing the data quality. The AUV can reduce the offshore
time by 40%.

CIM Cognitive Integrity

Management

Predict pipeline failures,

save lives, and protect the
environment ...
with the assistance of data science
and machine learning.

1-877-261-7045 onebridgesolutions.com/try-cim
Flow Forward™
new approach offers a more accurate and
cost-effective alternative to long-term evalu-
ation and prediction of a CP system perfor-
mance and overall pipeline condition based
on ROV surveys.
Case study
Due to the current state of the oil and gas
industry, clients are asking for ways to
improve pipeline surveys, from a technical
and financial standpoint.
A client recently reached out to
Oceaneering to inspect one of its pipelines
in the Gulf of Mexico. The client had used
an ROV for previous surveys. These surveys
had produced little useful data, and the client
was now interested in AUVs. It was fortu-
nate that the pipeline had existing data to
compare to that of the new technology being
implemented.
Over the coming years, the previous ROV
surveys will be compared to the ongoing
AUV surveys to determine if automation and
machine learning are providing the same
quality data as the ROV surveys.
The key performance indicator will be AUV
data as accurate, if not more accurate, than
the ROV-provided data. A faster AUV survey
is a requirement that is currently being met.
The key to using AUV technology coupled
with 3-D high-definition imaging is getting
rid of siloed information. Pipelines operate
over decades and information can be lost
easily. Additionally, technologies have made
some data incomplete or invalid. By overlay-
ing what may appear to be non-related data
sets, engineers are able to see patterns
that could not be seen previously, and this
enables better predictions, such as for the
loss of CP on a pipeline. However, AUV tech-
nology development is not without its chal-
lenges. For one, the computing technology
involved continues to evolve. OG
Adam Haavisto is an integrity engineer – offshore
asset integrity, Oceaneering.

FT4X Advanced Features:

‡ Robust 2nd Generation DDC-
Octanes Ethane Heptanes Sensor™
‡ Data Logger with date/time
Nitrogen Carbon Custom stamp & 40 24-hour daily
Dioxide Gas Mix totals
‡ Gas-SelectX® Gas Selection
Butane Methane Pentanes Menu
‡ Density and Gross Heating
Nonanes+ Hexanes Propane Value of Gases
‡ CAL-V™ Calibration
Set and Log a Custom Gas Mix! Validation
‡ Up to 1000:1 Turndown
The FT4X Gas-SelectX® allows the user to choose pure gases or a
custom gas mix with the gases available in the Oil & Gas and Mixed ‡ FM/FMc, ATEX, IECEx, and
Gas Menus. Set the gas composition in 0.1% increments in the field CE Approvals
and the meter’s Flow Calibration, Density, and Gross Heating Value are ‡ Optional RS485 Modbus RTU
automatically adjusted for the entered mix. The new Data Logger will store ‡ Free FT4X View™ Software
the gas composition and other configuration data.
Fox’s Model FT4X Data Logger records 40 24-hour daily totals, logs events and alarms with a 7 year history,
has settable Contract Time to define the contract day, and has a power off totalizer. The FT4X also has the 2nd
Generation DDC-Sensor™ and advanced correlation algorithms to provide accurate, multi-gas-capable gas
flow measurement. Direct mass flow measurement, exceptional low-flow sensitivity, fast response, and low
maintenance requirements also distinguish the Fox Model FT4X.
Email at sales@foxthermalinstruments.com, call us at
TRY THE
(831) 384-4300, or visit us online to find out how we can help FOX ONLINE
you meet your gas measurement and process efficiency needs. CONFIGURATOR
Use the online product configurator to customize a meter for TOOL

your application!
W W W . F O X T H E R M A L I N S T R U M E N T S . C O M Model FT4X Flow Meter

Challenge:
New U.S. Environmental Protection
Agency rules that go into effect in
2019 require refineries to quickly
and accurately measure heat
values in flare stack gases, and
typical stack analyzers fall short of
the accuracy and speed needed
to satisfy the EPA rules.
Solution:
To help refiners meet the new
EPA rules, AMETEK Process
Instruments committed to
developing a process analyzer
capable of measuring the heat-
ing values of vent gases, even in
widely changing flare gas streams.
Result:
The result of our development
efforts is the Dycor FlarePro —
a quadrupole mass spectrometer
specifically designed and
configured to provide rapid and
very accurate measurement of
flare-stack-gas BTU content.
AMETEK Dycor FlarePro™ Mass Spectrometer
Helps Refineries Meet New EPA Flare Stack
Heat Value Rules
Summary:
Determining the heating values of vent gas is important
because sufficient combustible material must be continu-
ously present to achieve high combustion efficiencies.
Many refiners already have flare stacks equipped with gas
chromatographs, but their slow, long cycle times make them
unsuitable for real-time control. Calorimeters, widely used
by refiners to determine fuel-gas BTU value, have limited
effectiveness in the presence of hydrogen and sulfur, which
typically occurs with refinery flare gas.
By contrast, field testing has shown that the Dycor FlarePro™
mass spectrometer offers more relevant, detailed data
faster than gas chromatography, while providing BTU-content numbers
equivalent to those from a calorimeter, but with greater specificity.
Using ASTM method D 2650-04, Standard Test
Method for Chemical Composition of Gases
by Mass Spectrometry, the FlarePro mass
spectrometer can monitor and quantity 16
components in flare gas streams, including
H2, N2, methane, ethane, propane, n-butane,
i-butane and pentanes, among others.
The resulting data is then converted into the
flare-gas heat value as described in ASTM
D3588-98, Standard Practice for Calculating
Heat Value, Compressibility Factor and Relative
Density of Gaseous Fuels.
The FlarePro also helps prevent smoking flares
by identifying such components as paraffin,
olefins, di-olefins and aromatics that raise smoke
index—a capability calorimeters lack.
Its proprietary software provides for easy setup, operation and calibration,
combined with advanced alarm and automation capabilities.
AMETEK Process Instruments
www.ametekpi.com
412-828-9040

Challenge:
In new gas pipelines, debris can
break loose and cause damage
to the meter’s probe. Getting a
replacement must be quick espe-
cially when it’s difficult to schedule
a gas sample analysis to be taken.
Solution:
The FT4A’s rugged DDC-Sensor™
is tolerant to debris in the pipe and
Fox can expedite an order in a week.
The FT4A allows the user to select
a custom gas mixture at the meter’s
display or a free software tool.
Result:
Production facilities see little to no
downtime by getting their meters
quickly. Selecting the gas on-site
takes pressure off timing of the gas
sample analysis.
BLM Regulations, Flaring, and the
Fox Model FT4A as the Solution
Summary:
The BLM has jurisdiction over most onshore leasing, explora-
tion, development, and production of oil and gas on federal
lands and each lease holder is responsible for maintaining
instrumentation to track “sales gas” flowing from the well site.
There are also rules for flaring waste gas at well sites.
Modern thermal mass flow meters are uniquely positioned
to solve these flaring measurement challenges as the
technology has recently seen advancement in accuracy and
the broadening of features.
1. Direct Mass Flow
2. Inexpensive solution
3. Accuracy and reliability
4. Ability to measure changing
gas compositions with
gas selection feature
In the harsh North Dakota environment, the
Fox Model FT4A Thermal Gas Mass Flow
Meter has been used at upstream oil
production sites in the Bakken.
A large Oil & Gas Company operating
numerous production facilities across
North Dakota reports that the Fox
Model FT4A has been introduced into four
key areas at their production facilities:
s test vessels for allocation purposes
s flare gas lines for waste gas or excess gas
s flare gas lines from tank batteries
s vent gas from tank batteries
Thanks to its robust DDC-Sensor™ design and innovative Gas-SelectX®
feature, the Model FT4A is making an impact in the gas measurement
applications in the Oil & Gas Industry while meeting the accuracy
requirements for BLM 3175.
sales@foxthermalinstruments.com
(831) 384-4300
www.foxthermalinstruments.com
EMERGING TECHNOLOGIES IN OIL & GAS

How machine learning

contributes to smarter
pipeline maintenance
Pipeline integrity management is an industry challenge, but a technology
breakthrough can help

L
By Tim Edward with Rob ast January, a major oil & gas company after excavating to investigate, crews began
Salkowitz ran routine inspections of its thou- sandblasting to clean the pipe. Moreover, the
sands of miles of pipeline, using the pipe ran through an almond grove, where each
same basic robotic device—the pig— tree is valued at more than $100,000. If the
that the industry has used for decades. new machine-learning solution had not discov-
However, this time, instead of sending ered the problem, the company estimated its
data from the pig to a roomful of analysts total exposure in the event of a failure at more
and waiting months for results, the company than $10 million. But with the right tools, they
applied a solution based on machine learning averted disaster.
Chart 1: Pipeline Incidents
and data visualization.
20-year trend. All graphics
The process identified a point of failure so A system at risk
courtesy: OneBridge Solu-
tions
severe that the pipe gave way as soon as, That’s just one story. Now consider that
there are an estimated 2.7 million
Calendar Total Cost Current miles of oil & gas pipes in North
Number Fatalities Injuries America, running next to factories
Year Year Dollars
and farmland, schools and hospitals,
1997 267 10 77 $110,377,793 mansions and nursing homes.
1998 295 21 81 $174,516,797 Nearly two-thirds of Americans live
1999 275 22 108 $178,313,209 within 600 ft. of a pipeline. And
2000 290 38 81 $257,659,464 in most cases, no one—not the
2001 233 7 61 $79,086,596 landowners, local authorities, or
2002 258 12 49 $124,085,321
even the company managing the
pipes—knows exactly where it is.
2003 297 12 71 $163,481,229
National regulators only require map
2004 309 23 56 $314,375,482 accuracies of plus-or-minus 500 ft.
2005 336 16 46 $1,477,030,849 Ask a crew chief how useful that
2006 257 19 34 $157,126,612 level of precision is when they are
2007 265 15 46 $147,823,421 out on a dig.
2008 278 8 54 $592,312,622 The Pipeline and Hazardous
Materials Safety Administration
2009 275 13 62 $180,372,220
(PHMSA) reported that significant
2010 264 19 103 $1,854,136,110
pipeline incidents grew 26.8% from
2011 287 12 51 $447,070,991 2006 to 2015. Those are incidents
2012 254 10 54 $233,822,619 that involved death or serious injury,
2013 304 8 42 $355,223,853 property damage in excess of
2014 301 19 94 $305,266,478 $50,000, or a fire or explosion. In
2015 328 10 49 $338,311,123 2015, there were 326 reported inci-
dents, nearly one per day.
2016 306 16 82 $301,612,864
No one understands this problem
Grand Total 5,679 310 1,301 $7,792,005,653 better than the industry leaders,
20 • APRIL 2018 OIL&GAS ENGINEERING
Figure 1: Before machine learn-
ing and pattern detection, sec-
tions showing greater than 25%
corrosion levels are highlighted
in red and only those areas were
excavated.

who have invested billions of

dollars and millions of hours
of manpower in efforts to
insure pipeline integrity. But
despite the best efforts of
dedicated professionals,
experts are fighting an uphill
battle against a relentless
foe: time.
Some 55% of liquefied
natural gas (LNG) pipes
in the US are more than
45 years old. Many of
them are quite a bit older
and far beyond their pro-
jected service life. Each
passing year puts more
systems at risk of failure.
Unfortunately, the tech-
nology to support integ-
rity management efforts
has not kept pace with
the scale of the issue, and that’s putting The task, however, is overwhelming.
intense pressure on the teams responsible Based on 25 years of field experience, one
for preventing incidents. of the authors of this article estimates that
How big is the gap between efforts and only 4% of collected data was being used
results? A recent Reuters report found to manage the pipeline, and even analyz-
that only 22% of 466 incidents tracked by ing that 4% is a gargantuan challenge.
Reuters were discovered using advanced Why are the numbers so low? Because
detection systems. Most were discovered the data is a huge pain to analyze. The pig
by members of the public who smelled gas produces a blizzard of numbers presented
or noticed a fluid leak. as a spreadsheet—a format that is now
more than 30 years old, and better suited
The data bottleneck to the days of green-screen terminals and
The problem with current integrity floppy disks than today’s era of intelligent,
management solutions isn’t the detection connected systems. It can take months
equipment. Pigs have earned their place just to break down one report, even under
at the front lines of defense in terms of the skilled eyes of senior analysts.
identifying anomalies. However, some Now imagine what will happen when
anomalies cannot be identified by current the industry’s most seasoned experts
technologies. And there’s no shortfall in retire. The skills they have are not easy
the effort and skills of the people tasked to to replace, and employers are unlikely to
analyze that data. Many have been on the tempt the rising generation of younger
job for decades and have developed the workers with a job that requires use of
insight and instinct to spot a problem in a archaic technology and a search for the
sea of data. proverbial needle in a haystack.
OIL&GAS ENGINEERING APRIL 2018 • 21

CUSTOM-DESIGNED
SENSOR SOLUTIONS.
PROVEN RELIABILITY.
At Pyromation, we’re dedicated to serving
the Oil & Gas industry with a comprehensive
line of RTDs, thermocouples and thermowell
assemblies. Our highly engineered,
custom-designed solutions meet industry
certifications and provide superior quality
control. Find out how we can meet your
temperature sensor needs.
GET A QUOTE!
pyromation.com/oilandgas
260.209.6341
NEC Rated · FM/CSA Approved
EMERGING TECHNOLOGIES IN OIL & GAS
Despite advances in computing technology, this has been
a hard problem to solve. Spotting specific patterns and
anomalies in a vast sea of barely structured data is a tough
task for computers that are great at making a lot of calcula-
tions fast, but not so good at tasks that require experience,
judgment, learning ability, and observational skills to visual-
ize real-world situations from columns of numbers.
Recent advances
Most recently, machine learning advances are helping to close
the gap between what people and what computers can do.
OneBridge Solutions has been applying emergent tech-
nologies to the problems of the oil & gas industry since
2014, building on Microsoft’s Azure: a cloud platform,
including services, for cloud-based development and
advanced data analytics.
Machine learning and the vast computing power of cloud
technology render the opaque mounds of pig data into
clear visualizations within a cognitive integrity management
SaaS solution. Features are aligned against the state of
the pipeline system, creating a new state, and then visual-
ized on a 3-D image replicating the actual pipe over the life
of the asset, versus run-over-run. With cognitive integrity
management, an entire pipeline system can be analyzed in
seconds or minutes rather than days or weeks.
Users drag-and-drop inline inspection tally sheets into the
solution, where it accurately ingests and normalizes them
into a standard taxonomy in minutes. Machine learning
allows for integrity management teams to gather insights
into the entire pipeline with ease and simplicity. New
recruits learn to use the system and analyze the results in a
much more intuitive, visual format.
It’s not just about speed. Increased precision and confi-
dence in data makes a difference in operational effective-
ness, efficiency, and enhanced threat detection. By deliv-
ering fast, complete, and accurate results each time the
pig is run, there is time to analyze beyond thresholds. An
anomaly growth can be tracked over time, before disaster
strikes. Rather than single data points along a pipeline,
interacting threats can be modeled.
Integrity managers needn’t fight a constant uphill battle
against data, time, and doubt. Decades of unreliable results
have led people to question corrosion depth data, even
on excavation. Now, with more precise ability to visualize
anomalies and validate results before digging, depth uncer-
tainty becomes a thing of the past. Crews armed with
better data move faster and cover more ground, with less
disruption at the worksite.
Getting the entire picture
Cognitive integrity management enables management
teams to gain insight and visibility into patterns and
trends that would otherwise be impossible to detect
22 • APRIL 2018 OIL&GAS ENGINEERING
Figure 2: As a result of machine learning and
the ability to ingest and analyze entire data
sets, corrosion patterns are visible along the
entire pipeline section.
due to the sheer volume of data
presented to the user. Pattern
detection provides some of
the greatest value to integrity
management since it unlocks to
power to optimize dig operations.
As shown in Figure 1, an integrity
team would have determined that
the areas highlighted in red were
the only sections along the pipe
with levels of corrosion that pose an
immediate leak threat. As a result,
these were the only sections of the
pipe that were repaired. Currently
within the industry, the thought
process is to respond to corrosion
growth and simply target immediate
threats rather than conducting a
comprehensive analysis of all threats,
both imminent and trending.
As shown in Figure 2, leveraging
machine learning and OneBridge
Solutions’ pattern detection algo-
rithm, the solution analyzed corrosion
growth along the entire section of
pipeline. When comparing Figures
1 and 2, corrosion was prevalent
throughout the pipeline. The sole rea-
son for the anomalies not being ana-
lyzed or detected when the pipe was
excavated was that corrosion levels
had not reached minimum threshold
levels. Now, through advanced tech-
nology, it became clear the entire
pipe was affected. Further analysis
determined that the corrosion growth
was due largely to faulty tape-coat-
ing, something that would have been
missed altogether in previous integ-
rity management practices.
Figure 3 shows the pipeline
through 3-D visualization, where the
pattern became much clearer and
more defined. Pipe zones, specific
target areas consistent with indi-
vidual threats, were identified and
searched. Corrosion growth associ-
ated with tape coat failure, in an aer-
obic environment, commonly called
“tape bagging,” is visible. The image
was enhanced by adding interacting
threat components, such as crack
and dent data. Further development
of this technology may advance the
detection of SCC
In the situation outlined above,
OneBridge Solutions was able to clear-
ly point-out scenarios where pipeline
OIL&GAS ENGINEERING APRIL 2018
23
EMERGING TECHNOLOGIES IN OIL & GAS

Figure 3: Cognitive integrity

management solution shows a
3-D visualization of the patterns
detected along the pipeline with
capability to drill down into spe-
cific areas along the pipe.

Machine learning pro-

operators could further optimize dig Finally, machine learning makes
management through greater visibility it easy for information officers and
into their pipes. With the advance- business decision-makers to say yes
ment of machine learning, pipeline to the tools integrity management
operators can be better equipped to teams need. This solution meets rig-
optimize their pipeline integrity man- orous enterprise standards for data
agement efforts. Dig management security, compliance, and manage-
becomes a straightforward task, with ability. It won’t tax the resources of a
the confidence to know that the entire stretched-thin IT staff. The goal is to
sections of pipe that are affected are make tedious data mining a thing of
repaired, rather than just sections that the past and free business leaders to
pose imminent risk. make intelligent decisions based on
solid data.
Put innovation in the pipeline
vides a way to give an
organization a competitive
advantage in the drive
toward zero failures. It
enlists the most power-
ful data and computing
technologies on the market
to help operators predict
pipeline failures, increase
safety, reduce operational
costs, and increase overall
confidence in the industry’s
ability to manage its assets
responsibly.
Leveraging the power of
machine learning, a costly
and damaging failure of the
pipeline running through the
almond grove was avoided.
But that’s just the start of a
journey where others across
the industry embrace smarter solutions
that address the looming crisis of main-
taining aging pipes. Modern cloud tech-
nology and emerging machine learning
solutions, delivered as a service, enable
industries of all types to embrace
an impending digital transformation.
Applying these technologies to pipeline
integrity management provides much
more: it unlocks the possibility of zero
pipeline failures. OG
Tim Edward is co-founder and president
of OneBridge Solutions, and a 30-year
veteran of the oil & gas industry. He has

Belt/Sheave
Every time a pipeline fails, lives, spent the past 22 years pioneering soft-
1-800-72-SHIMS (1-800-727-4467)

Laser Alignment System property, and the environment are at ware applications for the pipeline industry.
www.mrshims.com

New Green laser delivers these risk, and so is the reputation of the
important benefits:
● Reduces Vibration entire industry. For years, integrity Rob Salkowitz is an author, educator, and
● Eliminates downtime and productions
● At an affordable price
managers have been the last line consultant whose work focuses on the
● Visible indoors and Outdoors of defense, pitting their ingenuity, social and business impact of technology
● Brightness great for long distances
skills, and experience against a innovation. He writes regularly for Forbes
Mr. Shims rising tide of aging systems and a and serves on the faculty of the University
growing mountain of data.
your answer to better alignment for rotating machinery
of Washington CommLead program.

24 • APRIL 2018 OIL&GAS ENGINEERING

REFINERY OPTIMIZATION
Some alternatives for improving
an octane upgrade process
A properly optimized unit can recover 95% of the normal paraffins in the feed
By Robert Szczesnick
Figure 1: Octane upgrade
production and net mar-
gin. All graphics courtesy:
Honeywell UOP
A
s refineries seek efficiency, asset
utilization is a high-profile met-
ric. However, it is still common
to view efficiency and utilization
solely in terms of raw feed and
saleable product. Intermediate
processes, especially robust and stable pro-
cesses, are too often overlooked. Not evaluat-
ing a process because it is not failing or not a
marquee unit is a missed opportunity that can
lead to lost revenue.
Gasoline Molex provides an example of
such a process. An integral octane upgrade
process that feeds into the final gasoline pool,
gasoline Molex can be overlooked simply
because it is a reliable process that continues
to run with minimal attention. However, not
optimizing the process represents lost pro-
ductivity, efficiency, and ultimately, revenue.
Not only does an optimized gasoline Molex
unit produce more product in a more efficient
manner, but the increased product quality may
allow for operating conditions of lower sever-
ity in other process units.
Let’s look at some common gasoline Molex
operations scenarios and potential benefits of
optimization or revamp, including efficiency,
production, and maximum profit from existing
capital assets.
Robust and valuable
In the 1960s, Honeywell UOP pioneered the
physical separation of similarly boiling molecules
through the application of Sorbex technologies.
This concept was first used to upgrade the
octane content of fuels in 1990 with gasoline
Molex. Since then, gasoline Molex has proven
to be a valuable and robust component in many
refineries.
Gasoline Molex increases the octane value
of a light naphtha stream by separating lower
octane normal paraffins (extract) from other
components (raffinate). The higher octane raf-
finate stream is sent to gasoline blending, while
the extract stream is recycled, and combined
with a fresh feed stream, returned to an isomeri-
zation (Penex) unit. The isomerization unit prod-
uct is the feed stream to the gasoline Molex.
A new, properly optimized gasoline Molex
unit can recover 95% of the normal paraffins
in the feed, and typically returns these to an
isomerization unit in an 87% pure normal par-
affins stream. The removal of normal paraffins
increases the raffinate stream octane value,
while recycling the normal paraffins for
isomerization maximizes the value of
the entire stream.
If the recovery of normal paraffins
decreases, and more normal paraf-
fins slip out the raffinate stream, the
raffinate octane value decreases.
Additionally, normal paraffins bypassing
the recycle to the isomerization unit
represent a loss of potential octane
upgrade. As the isomerization unit
operates with an equilibrium reac-
tion, the end composition is relatively
unchanged, while feed quality affects
overall performance.
Feed to the isomerization unit with
higher normal paraffin concentration
OIL&GAS ENGINEERING APRIL 2018 • 25
REFINERY OPTIMIZATION
Figure 2: Adsorption section
production optimization.
26 • APRIL 2018 OIL&GAS ENGINEERING
has more potential isomerization of the normal
components or more octane upgrade. If the
extract stream’s normal purity decreases, this
represents a decrease in the potential upgrade
of the isomerization unit with additional circu-
lation costs.
Overall, at a constant feed rate, net margin
of a gasoline Molex unit is reduced 1% for
every 2% reduction in normal paraffin recov-
ery. Observations from typical units operating
for 10 or more years can be generalized as a
normal paraffin recovery decrease from 95%
to 90%, with a corresponding extract purity
decrease representing approximately 2.5% of
lost net margin.
Typical observations
Gasoline Molex ADS-35 adsorbent and
equipment are proven, robust solutions,
handling wide operational ranges, feed
changes, upsets, and other challenges
without permanent damage or reduction in
performance.
Gasoline Molex units frequently operate
for years without significant modification.
However, adsorbent ages and performance
factors change. These changes typically hap-
pen over a period of years. While different
engineers and operators transition in and out
of the operating unit, these subtle changes
may go unnoticed, though they do have
consequence. Thus, there is a need for care-
ful attention—which will be rewarded with
improved performance and profit.
Viewed in simple terms, the primary objec-
tives of a gasoline Molex unit are to deliver a
higher-octane product stream (raffinate) while
maintaining a highly efficient recycle stream to
the isomerization unit. As the adsorbent ages,
it is more difficult to maintain the performance
standards of 95% normal recovery and 87%
extract purity. Over the years, the highest
achievable values for these parameters may
begin to decrease.
At first, these changes are likely small and
within production tolerances. Eventually, a
single primary factor is maintain, while the
others continue to decrease. Provided the
minimum octane value is achieved, depending
on other unit specific factors, a common UOP
recommendation is to maintain the extract
purity and isomerization efficiency while the
recovery decreases.
As the recovery decreases, more nor-
mal paraffins slip through to the raffinate
stream. Octane values decrease with a
small increase in the flow rate. If produc-
tion is measured in octane-barrels (oct-
bbl), the loss is somewhat offset by the
flow increase. As performance decline
continues, raffinate quality decline out-
paces the volume increase with a mea-
surable decrease in oct-bbl production.
A typical response to decreasing oct-
bbl production is to increase the fresh
feed rate. This is possible, as decreasing
recovery results in decreasing extract
flow, opening capacity in the isomeriza-
tion and gasoline Molex units for addi-
tional fresh feed. Increasing the fresh feed
to the unit can appear attractive as raffinate
product oct-bbl values can be maintained, and
perhaps increased.
Initial results from this approach will sug-
gest operational adjustments have fully
compensated for changing unit performance.
However, the need for additional feed and
declining net upgrades from fresh feed to final
product are not considered.
First case study
Operations and flow rates in a gasoline Molex
unit can vary by significant amounts. This may
be due to the initial design process, different
feed compositions, feed availability, or a
variety of other factors. Frequently, multiple
variables change between two points of
comparison.
For an accurate assessment, a single fixed
operating reference must be established,
then variables adjusted one by one. Changing
individual variables is important to the gasoline
Molex unit as variables often have opposing or
compounding effects, making a change appear
more or less important than it actually is.
To establish a single reference point, a set
of primary operating characteristics were iden-
tified, and typical values were selected, such
as fresh feed rate, fresh feed composition,
and isomerization unit size. These values are
not intended to be average values, but reason-
ably represent most operating units. For this
evaluation, the typical fresh feed rate was
established at 65 T/h.
UOP’s UniSim yield estimate tool was used
to adjust the operating variables in a controlled
manner and estimate the results. This model
assumes the adsorbent chamber control sys-
tem (ACCS) and other operating parameters
within gasoline Molex are optimized, and the
unit is operating well.
The first phase examined the slow decline
of adsorbent performance with no adjust-
ments. Production simulation done at design
performance values established the reference
basis. As the recovery decreased, the raf-
finate product octane value decreased and the
extract flow (recycle to the isomerization unit)
decreased.
The net result was less material processed
through the isomerization unit as more normal
paraffins passed through the raffinate. The
reduction of overall oct-bbl and net oct-bbl
upgrade for a normal paraffin recovery of 90%
are shown in Figure 1.
The second phase increased
fresh feed to maintain design rates
for extract and combined feed to
the isomerization unit. With nor-
mal paraffins recovery constant at
90%, the raffinate octane value did
not improve, while raffinate flow
rates increased.
There was an overall oct-bbl pro-
duction increase compared to the
base design case. This flow rate
compensates for some of the per-
formance loss, but the net oct-bbl
upgrade is lower. The less efficient
upgrade of the additional feed
represents missed upgrade oppor-
tunity. These values are shown in
Figure 1.
Regularly scheduled optimization and per-
formance evaluations are best practices for
any operating unit to identify opportunities to
increase efficiencies and revenues. This is a
valuable exercise whether in the first year of
operation or after many years of operation.
Second case study
For example, a long-time operator of a
gasoline Molex unit with aged adsorbent
recently completed an adjustment of ACCS
zone ratio settings that increased production
more than 3% without increased feed.
ACCS zone ratios determine the chamber
circulation flows during a rotary valve cycle
and control the flows through the different
zones—adsorption, purification, desorption,
and buffer zones.
The gasoline Molex feed rate and other
operating conditions within the adsorption
and fractionation sections were maintained at
constant values. The ACCS zone ratio primar-
ily impacting the purification zone was adjust-
ed for evaluation while other ratios were held
constant.
Before the zone ratio adjustment, the unit
achieved a high recovery of normal paraffins.
However, the extract purity was low, nega-
tively affecting the Penex feed quality and
isomerization efficiency. The purification zone
ratio was increased, immediately increasing
the extract purity and isomerization efficiency.
Due to the age of the adsorbent, the high Figure 3: Alternative configu-
normal paraffin recovery and extract purity ration with one portion of
could not be maintained simultaneously. The the penetate diverted.
OIL&GAS ENGINEERING APRIL 2018 • 27
REFINERY OPTIMIZATION
Figure 4: Octane upgrade raffinate RON value decreased slightly with
protection and net margin. the minor increase of normal paraffin, but
remained within specifications at all times.
The net result was diversion of non-normal
paraffins from the extract stream to the raf-
finate product for an increase of 3% in octane-
barrel production with a constant fresh feed
rate. See Figure 2.
Third case study
When additional feed stock is available, there
typically is a good economic case to increase
unit throughput. Additional motivating factors
can include increased gasoline demand or
some combination of local market factors.
Typical primary issues with increasing
throughput are mechanical or hydraulic
limitations and capital. Through study of the
unit operations, changes are identified and
evaluated for maximum economic impact
with minimal investment. Typically, there
are multiple economically attractive options
identified to increase production with nominal
capital investment.
Larger mechanical unit components, such
as column trays, heat exchangers, or pumps,
are likely to yield reasonable incremental
throughput increase. An alternate flow con-
figuration can provide substantially larger
increases without the expense of building
new columns or reactors.
Consider an octane short unit with opera-
tional limits within the Penex reactor or frac-
tionation columns. A portion of the penexate
(Penex reactor effluent) may be diverted and
28 • APRIL 2018 OIL&GAS ENGINEERING
blended or stored as a prod-
uct, as shown in Figure 3. This
material has been isomerized
and the octane value has
been upgraded, though not
to the typical unit maximum.
Removing a portion of this
stream reduces feed to the
unit, which can be made up
by adding fresh feed to the
Molex feed.
This additional feed is
added until the extract flow
achieves the previously
established maximum rate.
The fresh feed has a higher
concentration of normal
components and the new
combined gasoline Molex feed contains less
non-normal paraffins.
While the extract and the combined feed to
the Penex reactor flows are not changed, the
raffinate flow rate decreases. The normal par-
affin recovery targeted, of 95%, is maintained
but the combined Molex feed composition
changes result in a different raffinate composi-
tion with a slightly lower octane value.
The net result is two product streams (high
and medium octane values) that can be main-
tained separately or combined. Combined,
this represents a 15% increase in feed capac-
ity, providing a 13% increase in oct-bbl pro-
duction. Results are illustrated in Figure 4.
Final words
A program of regular optimization and
performance evaluation can maximize
the value of operating assets. Honeywell
UOP’s gasoline Molex is a proven, robust
process that helps delivers long cycle time
with minimal changes. This longevity may
encourage complacency and lead to missed
value opportunities.
Investing time, resources, and capital,
however, can be rewarded with higher net
revenues and maximized returns on capital
assets. Fresh adsorbent, optimization, revamp,
or any combination of these investments may
be quickly repaid through higher efficiencies,
margins, and net revenue. OG
Robert Szczesnick is senior technical sales engi-
neer for detergents and olefins at Honeywell UOP.
 Save space and weight
With Precision Fluid Control
Products from Lee

Miniature Valves, Restrictors, Screens and Plugs for:

Innovation in Miniature
• Smartwell Systems
• Downhole Tools
• Subsea Production/Drilling Control Systems
• Subsea ROVs The Lee Company
NACE Compatible Products Available www.theleeco.com

Visit us at Booth 2109 at OTC

Westbrook • London • Paris • Frankfurt • Milan • Stockholm
INTRINSICALLY SAFE I/O...
Explosion Protection for Automation

Ex i System Highlights
r Approved for location in Class 1 Div. 2 / Zone 2
r Intrinsically safe [ia] digital, analog and safety modules
r Combine standard and Ex i I/O in the same node
r Modular, compact design – 12 mm width
r Fieldbus Independent

www.wago.us/hazloc