Copyright 2006, Society of Petroleum Engineers

This paper was prepared for presentation at the 2006 SPE Annual Technical Conference and
Exhibition held in San Antonio, Texas, U.S.A., 2427 September 2006.

This paper was selected for presentation by an SPE Program Committee following review of
information contained in an abstract submitted by the author(s). Contents of the paper, as
presented, have not been reviewed by the Society of Petroleum Engineers and are subject to
correction by the author(s). The material, as presented, does not necessarily reflect any
position of the Society of Petroleum Engineers, its officers, or members. Papers presented at
SPE meetings are subject to publication review by Editorial Committees of the Society of
Petroleum Engineers. Electronic reproduction, distribution, or storage of any part of this paper
for commercial purposes without the written consent of the Society of Petroleum Engineers is
prohibited. Permission to reproduce in print is restricted to an abstract of not more than
300 words; illustrations may not be copied. The abstract must contain conspicuous
acknowledgment of where and by whom the paper was presented. Write Librarian, SPE, P.O.
Box 833836, Richardson, TX 75083-3836, U.S.A., fax 01-972-952-9435.

Abstract
Slug flow, or instabilities, in multiphase wells and -pipelines
feeding oil- and gas processing facilities may cause
operational challenges related to avoiding unplanned partial-
or complete shutdowns, satisfying flaring volumes, managing
sand production, maintaining on-spec production, proving of
fiscal meters, etc. Conventional approaches to manage slug
flow in wells and pipelines include: choking the well/pipeline,
increasing gas lift rate(s), and/or providing overcapacity to
accommodate the slugs that are produced erratically. All these
conventional approaches imply inefficient production or
expensive over-design. In this paper, a differentfrom
choking and use of gas lift for stabilizationnon-intrusive
approach to slug management in wells and pipelines is
presented. However, it will typically be complementary to
other means, like choking and excessive use of gas lift, of
handling slug flow challenges. The approach also applies to
handling challenges related to oscillations and transients in the
total production system, not only in the wells and pipelines. It
may be interpreted as slug management from a practical
dynamical systems point-of-view. It has emerged as a result of
providing oil- and gas field operators with services and
solutions with respect to slug management since the late
1990s. The associated base of experience includes hands-on
experience from approximately 20 fieldsmainly in the North
Sea. The approach consists of solutions for automatic control,
or, active feedback control, of well(s)/pipeline(s) and
associated separators as well as services for generic handling
of oscillations/instabilities including control strategy selection
and -tuning. Field results from application of the approach are
included. The experience gathered during these years supports
that an integrated view of the dynamics of well(s)/pipeline(s)
and the processing facilities is required in order to sort out
proper activities and means for improving the slug
management. Furthermore, it is important that this view is
based on an analysis of relevant historical data. The analysis
phase should involve key personnel from operations like
senior operators, production-, and process engineers. It is our
experience that operators who, as part of their slug
management strategy, include an approach with characteristics
similar to the ones summarized above, will improve operations
significantly in an efficient and sustainable manner.

Introduction
Handling of oscillations, or instabilities, in oil- and gas
production systems is of high priority for operators of many
fields. The reason being associated profit-reducing issues like
unplanned shutdowns, too much flaring, increased
maintenance, off-spec production, metering problems, etc.
An oil- and gas production system consists of one or more
reservoirs, a production gathering system (wells/flow
lines/pipelines), processing facilities, export facilities,
associated instrumentation and controls, and a control system
with a certain configured control logic. Its dynamical behavior
depends on the combined state of all its components, or sub
systems, as well as how it is operated. Sorting out what is the
root-cause of oscillations present in the system, let alone find a
remedy, is therefore generally a very challenging task.
Tracking down prime suspects might in some cases be easy,
while fixing the associated operational problems might be very
hard and/or expensive. In other cases, it might be very hard to
find the cause, while the fix is easy. For example, spotting that
a pipeline or well is slugging, that is, producing liquids and
gas intermittently, is often easy based on appropriate pressure
and temperature readings. However, handling the problem
properly might be very challenging. On the other hand might it
be hard to spot that a level- or pressure control valve
experience stiction and moves in jumps causing
pressures/levels/flows/temperatures to oscillate. However,
fixing it might be easy. The same goes for the tuning of, for
example, a level controller.
In this paper it is presented a non-intrusive approach to
handling oscillations in oil- and gas production systems. When
it comes to slug handling, it might be complementary to other
more conventional non-intrusive means, like choking and
excessive use of gas lift for stabilization, or it might replace
them. In any case it enables increased production. It might be
interpreted as slug management from a practical dynamical
systems point-of-view. It has emerged as a result of providing
oil- and gas field operators with services and solutions with
respect to slug management since the late 1990s. The
associated base of experience includes hands-on experience
from approximately 20 fieldsmainly in the North Sea. The
approach consists of solutions for automatic control, or, active
feedback control, of well(s)/pipeline(s) and associated

SPE 96644
Active Slug Management
O. Slupphaug, SPE, H. Hole, and B. Bjune, ABB
2 SPE 96644
separators as well as services for generic handling of
oscillations/instabilities including control strategy selection
and -tuning.
The paper is structured as follows: Firstly, an overview of
the different activities associated with the approach is
presented. Secondly, overviews of associated control solutions
are presented. Before the conclusions at the end, some
example field results are provided.

Approach Overview
The approach to oscillation handling advocated herein may be
structured into three activities:
Data gathering
Data analysis
Proposition- and planning of field work
Data Gathering
The goal of the data gathering activity is to gather sufficient
data/information that the cause(s) for the oscillations can be
identified in the analysis phase. It is necessary to get a full
picture of the relevant part of the production system and its
associated control logic and operational procedures. It is often
very hard to know where to put the boundary for data
gatheringoften it is most time efficient to gather more than
possibly needed and discard what is not needed as the analysis
proceeds. The following data/information is typically
collected:
Logs of relevant tags (~#200-300) at 10 sec sample
rate
Data sheets (control valves, pumps, compressors, hx)
P&IDs
SCDs (System Control Diagramsif existing)
PFDs (Process Flow Diagrams)
Dump of control logic (from the control system)
Dump of database holding the controller settings
Control narratives
Overview of manual operations (the ones not
available in the datalogs)discussions with key
personell in operations
One lesson learned during the years is that the estimated time
for getting proper data logging in place and the proper data
into the office always is under estimated. An other observation
often made is that the control narrativesif existing at all
are very high level and does typically not include
considerations with respect to relevant disturbances to be
suppressed nor handling of possible loop interactions and
nonlinearitiesmain issues to be aware of to be able to
understand the observed oscillations. Yet another observation
is that it is very often inconsistensies found in the control
strategy shown in the P&IDs, the one shown in the SCDs,
and the one actually used as shown in the dump of the control
logic. Furthermore, is it very important to discuss the
oscillation nchallenges with experienced control room
operators. Their input is crucial in the analysis activity for
quality assurance of the data as well as explaining operations
not explicitly logged in the data. This includes opening and
closing of bypass valves, rerouting of wells, switching of
hydro cyclone packages, closing of block valves due to
leaking control valves, integrity testing of wells, cleaning of
transmitters, etc.
At the end of the data analysis activity a detailed view of
the relevant part of the production system, its instrumentation,
its control logic, and its operational procedures is in place.
This forms the basis for the analysis activity presented next.
Data analysis
In the data analysis activity a goal could be to provide a
prioritized list of remedies for handling the oscillations. A
different goal could be to answer whether a certain solution is
feasible and if it will cure the observed problems. In both
cases the result should be rooted in a proper, objective,
analysis of relevant historical data. The data analysis typically
involves analysing trends of hundreds of tags over a period of
a few months, typically sampled at 10 secs. Tools for doing
this efficiently are crucial. Furthermore, will the analysis often
involve carrying out dynamic simulations of the system. It is
then important that the underlying models are matched to the
historical data. Also, in order to configure a proper simulation
scenario, one need to fully understand the purpose of the
simulation including how to choose the boundary conditions
typically this requires extensive on-site practical experience.
The data analysis activity typically provides a list of
possible remedies, with associated cost and benefit estimates,
to the oscillation problems analysed based on historical data.
The proposed remedies can be categorized as follows:
Tuning of certain exising controllers
Change of control strategy
Active control of selected wells/pipelines
Active slug mitigation
Change of operating conditions
Change of operating procedures
Based on the above one carries on to the next phase if any
remedies have been found.
Proposition- and planning of field work
From the list of possible remedies one typically choose the
lowest hanging fruits first. This should be done in close
cooperation with operations so as to ensure ownership to the
chosen remedy in the on-site operations organization. This
typically gives a smooth and efficient implementation and
hand over of the remedy to operations. Also, it creates a basis
for efficient follow-up.

Slug Management Solutions Overview
As shown in Figure 1 solutions for slug management based on
active control may be structured into solutions for well
control, solutions for pipeline control, and solutions for slug
mitigation by utilizing buffering capacity in the processing
facilities. Common for all solutions are functionalities for
handling of non-linearities, decoupling of dynamic
interactions, as well as some sort of transient flow estimation
sufficiently accurate for control purposes. Next a brief
characterization of each solution is given.
Active Flowline Control
By active flowline control we refer to a control solution which
actively manipulates the flowline outlet control valve or choke
based on properly chosen measurements in the pipeline and
the downstream processing facilities. Typically, the main goal
is to provide operations with the possibility of operating at the
highest possible average valve/choke opening or at the lowest
possible wellhead platform (or, subsea manifold) backpressure
without causing shutdown or overload of the downstream
SPE 96644 3
processing facility. A possible layout for so-called active
flowline control is shown in Figure 2.
Active Well Control
Active well control is basically active flowline controls
counterpart for control of wells and by active well control we
refer to a control solution which actively manipulates the
wellhead amd/or gas-lift choke based on properly chosen
measurements in the well and the downstream processing
facilities. Typically, the main goal is to provide operations
with the possibility of operating at the highest possible
average choke opening or at the lowest possible wellhead- or
downhole pressure without causing shutdown or overload of
the downstream processing facility.
Figure 4 illustrates the different functionalities included in
the solutions for active flowline control and active well
control. As can be seen there are modules for stabilization of
the flow in the pipeline/well, avoiding overload of
downstream facilities, support of ramp-up, pro-active slug
reception, and monitoring of the well/pipeline state.
Active Slug Mitigation
Active slug mitigation refers to a control solution which
mitigates flow variations through the processing facilitites.
This requires controlling at least one of the inlet separator(s)
outlet valves. It is typically integrated with active well control
or active flowline control. This way one gets an integrated
well/pipeline separator control strategy, meaning that one can
coordinate the control of the both the inlet and the outlet of the
separator so as to increase the maximum possible average
pipeline/choke opening or minimize the possible associated
backpressures even more than with out such an integration
(while avoiding overload of the separator and possibly other
downstream equipment). An example layout is shown in
Figure 5.
Field Examples
We have included three field examples to illustrate the
potential benefits of the above described approach and its
associated solutions. Common for all these examples is a very
tight cooperation with the on-site operations team as well as
thorough preparations based on historical data.
Field Example 1
Figure 5 and Figure 6 basically documents the effect of active
slug mitigation combined with active flowline as advocated
herein compared to a different slug mitigation scheme with the
same goal of maximizing the average pipeline outlet choke
opening.
Field Example 2
Figure 7 and Figure 8 document that it is indeed possible to
stabilize the flow of well using active well control based on
the downhole pressure.
Field Example 3
Figure 9, Figure 10, and Figure 11 illustrate that proper tuning
of a degassing drum level controller can increase the water
treatment capacity of a processing facility by 6%, which was a
major benefit since it was constraining the production. In this
case it is interesting to note that the main focus was on the
wells and the inlet separator control in the beginning of the
analysis.
Conclusions
The presented non-intrusive approach to handling oscillations
in oil- and gas production systemswhich might be
interpreted as slug management from a practical dynamical
systems point-of-viewhas proven to enable substantial
increased production using the different associated means.
Success criteria include; (1) thorough preparations in form
gathering data and information so as to get a complete picture
of the production system with its associated actual control
logic, instrumentation, and operational procedures; (2) data
analysis using proper tools and with involvment from
experienced personell from operations for quality assurance;
(3) planning of on-site work realizing the potential of the
lowest hanging fruits first, again with involvement from
operations.


Figure 1 Different solutions and services associated with a non-
intrusi ve approach to slug- and oscillations management.

L
P
z
Separator
P
P
F
AFC
P

Figure 2 Possible layout for acti ve flowline control. The flow out
of a pipeline tying a wellhead platform to a processing platform is
controlled. The measurements used are the wellhead platform
departure pressure, the pressures up- and downstream the
pipeline choke and the pressure and oil level of the receiving
separator. Based on these measurements the choke is
continuousl y acti vated so as to attempt stabilizing the flow in the
pipeline while avoiding overloading the downstream separator.

4 SPE 96644

Figure 3 Possible layout for active well control of a gas lifted well.
Measurements used are the gas-lift annulus pressure, the
downhole pressure, and the pressures up- and downstream the
production choke.


Figure 4 Functionality overview for acti ve flowline- and well
control.


Figure 5 Field Example 1: Acti ve slug mitigation. Showing
measurements (blue) and controls (red).

Blue Inlet choke
Red Oil flow
Yellow Oil level
Switching control strategy

Figure 6 Field Example 1: Comparison of two different control
strategies for slug mitigation. The one to the right is the one
recommended herein. In this case it resulted in a 3.5 % increase in
liquid production compared to the one in place at the time of the
trial. In addition the separator oil level excursions were
significantl y reduced.


Figure 7 Field Example 2: Acti ve well control. Instrumentation and
well geometry.

2 2.2 2.4 2.6 2.8 3 3.2 3.4 3.6 3.8 4
74
76
78
80
82
84
86
88
Time [days]
Brage Active Well Control Field Test (Day 0 =24-Aug-2001 07:59:00)
Starting active control
Days

Figure 8 Field Example 2: Result of acti ve control of the well.
Downhole pressure is stabilized and reduced in a controlled
manner. In this case the well was impossible to produce with out
the stabilizing controller.

SPE 96644 5

Figure 9 Field Example 3: Initial distribution of flow rate through
degasser.


Sammenheng mellomventilpning og vannrate, ventil til sj
y=0,003x
3
- 0,1749x
2
+4,8379x- 5,2632
-200
0
200
400
600
800
1000
1200
0 10 20 30 40 50 60 70 80 90 100
0209- 0409
2308- 2408
Poly. (0209- 0409)

Figure 10 Field Example 3: Installed control val ve characteristic.
Gain is increasing with increasing val ve opening and throughput.
X-axis is val ve opening, y-axis is throughput in m3/h. PID
controller tuning at low openings results in oscillatory behavior at
high openings.

Figure 11 Field Example 3: Final distribution of flow through
degasser. The flow variance reduction enabled a 6% increase in
the constraining water treatment capacity since the acti ve
constraint was on the pressure safety val ve flow capacity for the
degasser.