SPE 103082
Intelligent Well Automation—Design and Practice
D. Mathieson, SPE, C. Giuliani, SPE, A. Ajayi, SPE, and M. Smithson, WellDynamics Inc.
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., 24–27 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
The commercial application of Intelligent Well Technology is
fast approaching the ten year milestone. Expanding the range
of advanced completion and monitoring products with
particular focus on reliability engineering has been the main
theme for this sector. As this technology has matured,
research focus has shifted to Intelligent Field or Smart Field
concept studying the potential of wide-scale automation and
optimization.
This paper describes the development of Intelligent Well
surface control systems, aimed particularly at providing
automation and optimization of well based processes as a
means to achieve advanced well management. A simple layer
model is proposed describing an increasing level of control
and automation.
Data standards and connectivity are discussed as the
vehicle to achieve wide-scale integration of advanced well
management, as this capability is portrayed as one of the
stepping stones of Intelligent Fields.
Well Based Processes – The starting point
An Intelligent Well (IW) provides the means to control the
process of producing hydrocarbons from or injecting fluids
into the reservoir. There are three elements to performing
control: process outputs, controlled variable and manipulated
variable. The process outputs are those parameters selected to
describe the state of a system and are monitored continuously
by sensors (process observability). The controlled variable is
the parameter through which the process outputs are controlled
(e.g. zonal pressure, flow, water cut, etc) and the manipulated
variable is the parameter of the system used to control the
process (e.g. valve position). The relationship between the
controlled and manipulated variables is defined by the control
algorithm (supported by static inputs to define and calibrate
the control model in the decision space) and the control
strategy implemented. The implication is that the downhole
valve flow trim and the accuracy and resolution to position the
valves are the fundamental parameters to deliver effective
control of the well process.
The flow control valves can be divided into two main
categories with respect to their control capability. There are
binary and multi-position valves. A binary valve provides two
setting options for the manipulated variable: open or close
position. A multi-position valve provides a much wider range
of options for the manipulated variable and can be further
divided into discrete or continuous type depending on the trim
design. The discrete valve allows for stepped changes of
available valve positions. The number of finite positions
depends on application specifics and sleeve operating
philosophy and mechanics. Continuous valves allow for
infinite position placement within the prescribed flow area
boundary when mechanically coupled with an adequate
control system; this design provides a finer resolution to the
control process. Continuous valves can also be deployed with
a control system that delivers less accuracy and resolution in
position placement. This solution can be defined as
continuous multi-position.
A binary valve is appropriate in situations where the
objective (or driver) is the exclusion of a zone when the
production of unwanted fluids (water or gas) cannot be
managed once the breakthrough occurs. The downhole flow
control valve would be closed position thus resulting in loss of
production form the associated zone.
The discrete multi-position valve is most suitable in
applications that are required to meet changing control
objectives such as gas lift, where the amount of gas injected
can be adjusted as function of produced fluid composition,
available injection and reservoir pressures, or in injection
wells where the reservoir sweeping efficiency can be
improved by optimized allocation of injection fluids.
The continuous infinite valve is usually beneficial in
applications that require a tighter control. A WAG well,
requiring the control of both water and gas injection) is a
typical example alongside management of gas problems in oil
producers and controlled pressure drawdown at the wellbore
and sand face completion.
Field Management Requirements
When the optimal solution for controllability (possibility to
force the process into a particular stage) and observability
(ability to “observe” the process through output measurement)
of the IW system is established, the design of the IW surface
control and downhole infrastructure should address the
operational philosophy at field level. While project specific
2 SPE 103082
constraints are numerous, those that should be at the forefront
of any applications are:
• Requirement for on-line data and control
• Constraints from existing field infrastructures
• Automation, control and optimization capability
On-Line Data and Control
The streaming of data from field sensors to a variety of
monitoring systems, ranging from alarms servers and graphing
products to the more sophisticated expert systems, define the
concept of online data. This capability is critical where early
detection or prediction of process disturbances is required.
Similarly, on line control, defined as the ability to control the
IW process on-demand through existing field infrastructure, is
best deployed in those applications where the time between
the decision to intervene on the well and the intervention must
be kept to a minimum. This feature is desirable when control
loop ref 1, automation or remote operation is a requirement.
Requirement for on-line data and control may be driven also
by accessibility of the well site or safety concerns.
The solutions for data and control may not always be
integrated; an on line data system could be combined with a
campaign of well interventions carried out using a portable
control system unit. However, more often a random sampling
(i.e. field operator traveling to the well site according to a
predefined schedule) of the well metrology is implemented
alongside the portable unit. Both solutions are characterized
by considerable lag time between the occurrence of the well
process disturbance and the needed modification to the well
process with the intervention (i.e. manipulation of downhole
flow control valves). Such design of IW control system is
suitable for those applications where no sudden or frequent
variations to the well process are requested or expected. Other
reasons relate to limited field infrastructure or excessive cost
to put infrastructure in place.
Existing Field Infrastructure
Historically, when referring to the field infrastructure,
power and RTU communication standards were the only
criteria required to design a surface system. While they are
critically important for a specific application, additional
factors such as field control system implementation
philosophy, field historian accessibility and connectivity
between the process and the business networks are now as
important in the definition of the design criteria for IW control
systems (fig. 1). As IW are deployed to mitigate uncertainty
and risks or manage complex downhole architectures, field
communication infrastructure should always be designed as
open architecture ref 2 in order to both facilitate late integration
of critical subsystem and better allocation of the system
functionality among the different field processes. Field
communication infrastructure that does not allow for open
architecture may force the IW control system to be embedded
into the field control system.
Integrating the IW control system into the field process
network as a sub-system, allows the IW capability to evolve as
new technology reaches the market; this is because the
connectivity with the field control system occurs at the
application layer instead of the data layer. Embedding, on the
other hand, often freezes the development as the field control
system providers may have little interest in the implementation
of other service company products.
Automation, Control and Optimization
It is important to identify the requirement for automation,
control and optimization in the different processes.
Automation, for instance, can have several applications in an
IW control system. For asset managers such as Production
Engineers, automation can be applied to execute pre-defined
sequences of operations triggered by user defined events ref 1
(e.g. start a pressure build up when a downhole valve is
closed).
To the IW completion design, automation delivers control
of downhole flow control valves and data acquisition from
deployed sensors. As flow control valves act as the
manipulated variable to control the well process, it is obvious
that the repeatability and consistency of operation delivered by
automation is at the foundation of any advanced well
management control and optimization capability.
Advanced Well Management algorithms, reservoir
performance, downhole valve profile and automated operation
of downhole flow valves are all part of the design inputs for
the IW control system and its surface components. How they
interact to deliver control is simple: advanced well
management (e.g. well output) establishes the set point for the
controlled variables (e.g. zone pressure, zonal or total well
flow, water cut, etc) and IW process controller converts this
input into the required downhole valve position and the
automated operation of the manipulated variable takes place.
As the relationship between the controlled variable and the
manipulated variable is specific to the design of the downhole
valve flow trim ref3, it is reasonable to implement any well
process automation closely coupled to the IW control system
or, for simpler tasks, embedded into the IW control system.
Optimization algorithms to enhance the performance of the
system would follow a similar pattern. The optimization
module generates the set point for several processes within the
field concurrently, with one or more related to the process
controlled variable. Most of the optimization functions
connect the business requirements of the Operator to the
production process and have space application and modeling
demands that go beyond the capability of an IW Control
System and are implemented in the business network (Fig.1)
Surface Control System Design Philosophy
The Surface Control System is an extension of the
downhole IW control system architecture. The great majority
of IWs deployed to date rely on locomotive power to be
delivered from surface to a downhole flow control valve by
means of hydraulic or electrical conduits. The addition of
downhole mechanical or electronic devices may be necessary
to deliver an increased control over the flow control valve
movement. These elements constitute the downhole control
system architecture. Similar downhole control methodology
on the market follow a simple classification:
Direct Hydraulic, a system based on direct connectivity
between the source of pressurized hydraulic fluid at surface
and the downhole actuator. This system requires a minimum
SPE 103082
number of control lines emerging from the well equal to the
number of downhole devices to be operated hydraulically.
Positioning of the valve may depend on downhole mechanical
devices or surface equipment.
Multi-drop Hydraulic, a system based on sharing a
reduced number of control lines among several downhole
actuators and selectively establishing the connection between
an actuator and the source of pressurized hydraulic fluid. This
system requires the number of control lines emerging from the
well to be less than the number of downhole devices and it is
most suited for those applications where well completion
architectures do not allow for numerous control lines.
Positioning of the valve actuator may depend on downhole
mechanical or surface equipment.
Integrated Flow Control and Monitoring, a system based
on integrating flow control and sensing into the same power
and communication infrastructure, capable of selectively
operating one or multiple actuators through the available
locomotive power. Positioning of the valve actuator is
typically very accurate with a high degree of resolution and
obtained with a close-loop control implemented within the
downhole electronics.
Downhole Monitoring, is often used in combination with
surface metrology to acquire feedback on the well production
or injection processes after any of the manipulated variables is
adjusted. Established permanent downhole gauges (PDGs) are
primarily based on electronic or fiber optic systems. They
allow monitoring of similar well process parameters such as
pressure, temperature, flow and some aspects of flow
composition. Electronic based system may be preferred for
applications where valve position and high accuracy and
resolution for pressure and temperature measurement are
critical. Fiber optic, on the other hand, is more effective in
distributed measurement (pressure and temperature).
Dedicated downhole monitoring is typically used in
combination with hydraulically operated systems to provide
comparable capability to the integrated flow control and
monitoring systems. It has an advantage of greater system
flexibility as well as improved mission reliability.
Control and Automation – Layer Model
The manner in which downhole sensing and flow control
systems are combined and deployed to control the well
process is at the base on the layer model. The degree of
control with which the production engineer intends to manage
the well process should dictate which level of functionality of
the downhole control architecture and associated surface
equipment. This may also require connecting an IW system to
other well based systems such as ESP, surface choke and gas
lift systems ref 1. The following classification of process
control level is proposed for the purpose of this article:
• On / Off
• Discrete Multi-Position
• Continuous, low resolution and accuracy
• Continuous, high resolution and accuracy
Manual Systems, typically designed for hydraulically
operated downhole control systems, rely on the field
operator’s knowledge of the downhole control system to
3
manipulate the flow control valves to the required position and
its operational feedback to confirm the new position. These
systems supply the locomotive power necessary to operate
downhole actuators and provide a method to manually distribute
it to the downhole actuators. Regardless of the system being
portable or permanently installed in the proximity of well, the
operator is required to travel to the well site.
Manual systems do not typically include a data acquisition
system for downhole sensors. If downhole metrology is
present, it would be available to the operator as external
feedback onto which an assessment of the operation success
should be made. Manual systems provide the lowest level of
control capability and are typically associated with On / Off
and Step control.
Automated Systems, are designed for repeatability and
consistency of downhole flow control valve as their primary
objective. The operator is no longer required to be intimately
familiar with the details of the downhole control system
operation and feedback, but can shift his attention to the well
performance. Automated systems are designed to operate
individual well with specific downhole architecture and would
comprise similar basic equipment to the manual counterpart.
This is the entry level for advanced well management.
Automated systems often include or are connected to data
acquisition systems from which they can obtain feedback on
the process being controlled. This feature enables automated
systems to contain simple well process automation and control
loops which do not require complex modeling.
Automated systems can either be deployed as permanent or
portable units. Permanent solutions are typically connected to
the field communication infrastructure and are available on
line for remote operation. Portable units, designed to be easily
transported to the field, are normally operated on site (offline)
or remotely if an ad hoc communication network with the
control centre is available or established at the well site. When
automated systems are deployed offline data can still be
streamed into expert systems or, in most cases, analyzed after
intervention.
IW Supervisory Systems, deployed in conjunction with
automated systems, are designed to bring together supervision
and control for all IW completions deployed as well as
external connectivity.
For field development where the IW completions are
critical elements of the field exploitation strategy, an IW
supervisory system has the significant advantages:
• Provides integration at application level for easy data
and control transfer to and from other field control and
asset management systems
• Provides a deployment base for computational
intensive well management capability and advanced
automation sequence spanning multiple processes
• Facilitates the introduction of modification to hardware
and control philosophy to any IW system deployed in
the field
• Relieves field control system providers from
developing an in depth understanding of IW control
system and control strategy
• Allows any combinations of downhole architectures
and functionality
4 SPE 103082
• Provides Well centered data processing and data
storage ref 4
IW Supervisory Systems offer the most in terms of IW
functionality packaged into a single product.
From Intelligent Well to Intelligent Field
The Intelligent Field concept integrates all the systems that
affect the recovery of hydrocarbons from source to delivery to
the customer. From reservoir, through the wells, artificial lift
systems, the gathering systems, manifolds, separation and
upgrading systems, to the products custody transfer and
disposal of effluents, the Intelligent Field interconnects and
brings together a number of processes. The objective of the
Intelligent Field technology is to improve the efficiency,
individually and collectively, of all the above processes. This
is not likely to be achieved by the construction of massive
monolithic process model, control system and data storage
which embeds all of the necessary functionality, but by the
integration and interaction of each of the main systems which
constitute it. Earlier in this article, the case was made for IW
control system with increased levels of control and automation
ranging from manual systems to advanced well management.
The philosophy of integrating IW into Intelligent Field should
reflect the criticality of managing the well process in the
context of the field exploitation plans.
For field applications where the IW is utilized to optimize
production, measure effluents or automate well surveillance
processes or any other task which is computational intensive
or requires specialized knowledge of advanced well
management, the IW control system is one of the systems
constituting the backbone of the Intelligent Field solution.
The process control infrastructure is such that the Intelligent
Field would see the IW Control System as a peer to the Field
Control System (Fig. 1).The process network provides
concurrent access to in-field well controller controlling the
different well systems such as IW, ESP, Gas Injection and
Surface Choke to name a few. The field control system
supervisory application has access to IW, downhole
monitoring and any other functionality available through the
IW well controller through the IW Supervisory Systems.
Conversely, the IW supervisory application is able to define or
“recommend” to the field control system operator new settings
or notify unplanned operating conditions. A similar solution
can be proposed for monitoring only application that require
specialized and computer intensive calculations.
For less calculation-intensive process automation, effective
downhole system control strategy or well integrity related
solutions, automation must be deployed at the well site. A
local communication network allows for the different well
controllers to exchange data upon which automated process
can be implemented. This solution would not negate the
requirement for supervisory system but would remove any
uncertainty related to well site to control room communication
infrastructures.
In cases where Intelligent Wells are installed with the sole
purpose to reduce intervention time and cost, the IW control
system is easily embedded into the field control system and
used only as a mean to manipulate the valves.
Integration to Intelligent Field
Automation and control loops have been implemented to well-
centered processes for quite some time in control systems for
surface chokes, surface gas injection valves and ESP systems.
Operator familiarity with the IW completion technology has
increased to the point that automation and control loops may
be introduced in IW control systems ref 5 with a plan to address
stability and safety.
Key to achieving a wide scale application of these
advanced well management capabilities is the ease of
integration into field control systems and Intelligent Field.
This can be achieved with open architecture and adoption of
industry standards.
TCP / IP, is one of the emerging standards for process
control networks. This system supports several data protocols
on the same physical layer, although it does not provide
inherent interoperability between application layer protocols.
However, TCP / IP allows for in-field controllers from
different sub-system on the same network making expanding
the communication infrastructure relatively easy (if bandwidth
for data throughput exist). This capability is also available in
wireless technology, though with limitations.
OPC, in recent years, has gained considerable market share
as application layer protocol to exchange data. Built on
DCOM technology, OPC does show some limitation on
complex network or crossing gateway between business and
process networks. This technology is evolving towards a new
standard, OPC UA (Unified Architecture), which will address
the above issues.
PRODML, The upstream oil and gas industry is actively
working to establish standards for application connectivity.
PRODML (PRODuction xML) is an industry initiative whose
objectives are to achieve data and control interchange using
XML as underlying technology (www.ProdML.org).
Conclusions
Control system is one of the IW elements key to a successful
deployment of IW technology and the future of Intelligent
Field. It links and integrates the process control capability of
the IW downhole hardware to the data analysis and decision
making processes as well as the intervention phase.
The starting point of designing an IW control system
should be a clear understanding of the degree of control
required by the well process and whether enough parameters
are monitored to provide observability to the system. Control
specifications should then be set to make certain that the
process controller model selected provides the necessary
stability and robustness with the selected control strategy. A
map of systems functionality and interfaces should also be
drawn up to best determine in which componet (e.g. in-field
controller, supervisory software, business network)
automation and optimization algorithms should be
implemented.
For some applications, most of these steps are not
necessary because limited control is required. More generally,
most of these steps are not implemented as the IW control
system is not an integral part of the IW system design. The
authors concur with the vision that there are no technical
barriers ref1,5 to the implementation of automated processes
including control loops. However, it is important that
SPE 103082 5

communications infrastructures and the field control system

philosophy allow for data exchange through the relevant
systems at either in-field controller level or application level.
The challenge is to bring the decision on IW further forward in
the field development plan to make sure that infrastructure are
designed with this requirement in mind.
An IW system, modular in hardware and software design
and based on open architectures and industry standards, will
enable deployment of automation and control capabilities in
most field control system architectures, thus delivering
advanced well management modules fully integrated with the
Intelligent Field technology.

Nomenclature
IP Internet Protocol
OPC OLE for Process Control
OLE Object Linking and Embedding
RTU Remote Terminal Unit
TCP Transmission Control Protocol
WAG Water-Alternate-Gas

References
[1] Going, W.S., Anderson, A.B., Vachon, G.P.: “Intelligent Well
Technology – The Evolution to Closed-Loop Control,” paper
OTC 17796, presented at OTC Conference 2006, Houston,
Texas, 1 – 4 May
[2] Hiron, S.: “Networking Subsea Completion Using Industrial
Standards,” paper SPE 71532, presented at 2001 SPE ATCE
2001, New Orleans, Louisiana, 30 September – 3 October.
[3] Konopczynski, M. and Ajayi, A.: “ Design of Intelligent Well
Downhole Valves for Adjustable Flow Control,” paper SPE
90664, presented at SPE ATCE 2004, Houston, Texas, 26 – 29
September 2004
[4] Leonard, S.:”MatrikonOPC Hub and Spoke Architecture”, white
paper
[5] Going, W.S. et all: “Intellegent Well Technology: Are We Ready
for Closed-Loop Control?, paper SPE 99834, presented at SPE
Intelligent Energy Conference 2006, Amsterdam, The
Netherlands, 11 – 13 April
6 SPE 103082

Business
Historians Asset User
Software
Managment Interface
Solution

Business Network
Security
Gateway

User
Interface

Intelligent Field
Fiscal Safety
Metering
Well Control
System
Control System
Process Network
TCP / IP

Intelligent Wells
Field Field Surface Choke
ESP
Devices Controllers Gas Injections
Chemical Injection
etc
Figure 1 – Business and Process networks connectivity overview