NEWS

NEWS

Reservoir Evaluation
Production Optimization
Simulation
Well Testing

Software

Winter 2002

UPGRADE

Engineering and Business News From Fekete Associates Inc.

F.A.S.T. WellTest TM users will be

impressed with the newest version of the
software to be released in February 2002.
Included on the list of upgrades are:

New model interface with on-screen

reservoir schematic that displays all
boundary conditions as they are
selected. The schematic shows
reservoir dimensions and distances to
no-flow/constant pressure boundaries
in both the horizontal and vertical
directions.
Enhanced parameter editing
functions, which include default
selections from existing analyses.

Improved production forecast graphs

with any combination of cumulative
production, reservoir pressure and
flowing pressure versus the production
rate. Oil forecasts display reservoir
pressure decline above bubble point.

New model for hydraulically fractured

wells with boundaries.

New model for a well in a wedge.

Radial composite model with

unlimited regions.

A multi-layer model with unlimited

layers communicating at the wellbore.

Enhanced plot customization.

The Rate-Pressure Editor (formerly

called Production History editor)
has been modified to facilitate the use
of Date/Clock Time. Afterflow rates
are now calculated and displayed. The
graphical rate editor has been modified
to allow more flexibility in entering
production data.

Fekete Associates Inc. And Rocky Mountain Veco

Announce Alliance
Fekete Associates Inc. is excited to announce
an alliance agreement with Rocky Mountain
Veco of Englewood, Colorado. The alliance
was formed primarily on behalf of Feketes gas
modeling group and Vecos upstream oil
and gas engineering group.
Using information generated
by a Fekete pipeline model,
Veco engineers can design and
construct pipelines and
facilities to match the deliverability of a
gathering system through time. Direct
communication between Feketes
modeling engineers and Vecos
design engineers will lead to
savings in both time and cost.
Feketes engineers use the
F.A.S.T. Piper software,
developed in-house at Fekete, to
conduct modeling and optimization
studies for gas producing clients. F.A.S.T.
Piper is currently used by over 400
companies throughout North America and
internationally. This year, the Piper group

Save Time and Money!

Avoid the hassle of organizing
and forwarding well test data.
Your wireline, testing and
laboratory suppliers can now
submit the data directly to Fekete
via our secure website. You get a
notification that your data has
arrived. See our website
www.fekete.com to set up a
free account.

has conducted studies throughout North

America, Australia, and Pakistan on over 3000
conventional and coalbed gas wells.
Recommendations include looping pipelines,
adding compression, and adding
facilities. The impact of upcoming
drilling programs are
evaluated to maximize
production return from new
and existing wells.
Rocky Mountain Veco is a full service
engineering design, procurement,
and construction company.
Projects include arctic, frontier
and offshore gas gathering
system design, oil and chemical
processing facility design, and
various related construction
projects. Valuable experience
has been attained through over 30
years of design and construction
experience in 40 countries. For more
information on Rocky Mountain Veco, visit
their website at www.veco.com or contact
Larry Compton at (303) 792-2211.
For more information about how this
agreement will benefit you, or to start
streamlining your pipeline and facilities
operations with our full service modeling
capabilities, contact Kevin Dunn at (403)
213-4200 or toll-free at 1-800-625-2488.
INSIDE THIS ISSUE
Veco/Fekete Alliance

Software Upgrades

1&4

Tech Talk:
Advanced Decline Analysis

2&3

Fekete on the Road

For details on all our software products and engineering services, visit us on the web at www.fekete.com

Tech Talk: Get More From Your Produc

Louis Mattar, MSc, P.Eng., is President of Fekete
Associates Inc. He has authored more than 40
papers, and leads the software development group
at Fekete. Louis co-authored the ERCB Theory &
Practice of the Testing of Gas Wells in 1975, an
authoritative text on the subject. He has appeared
as an expert witness in regulatory hearings.
Engineers use production data to draw conclusions about reservoir size
and quality, and to identify operational problems such as liquid loading
in the wellbore. Production data is plentiful, easily accessible, and
inexpensive to obtain, which makes it an ideal source of reservoir
information. All reservoir engineers are familiar with the concept of
simple decline analysis, which attempts to fit empirically derived curves
to daily or monthly production data to forecast future production and
predict recoverable reserves. The attractiveness of this method lies in its
simplicity, but unfortunately, the results that we obtain from it can
sometimes be misleading. This arises from the fact that basic decline
analysis techniques tend to ignore half of the information, namely the
flowing pressure.
There is much more information to be gleaned from your production
data, if you know where to look. By including flowing pressures in your
analysis, and by using more sophisticated decline analysis techniques,
increased accuracy in reserve estimates can be easily obtained, and
pitfalls can be avoided. A Material Balance analysis can be performed
without ever having to shut-in a well. In addition, early-time production
data can be used to determine reservoir quality and stimulation
effectiveness.

A Material Balance Analysis can be performed

without ever having to shut-in a well.

David Anderson, BSc, P.Eng., is the lead

development engineer on the F.A.S.T. RTA
software. He has conducted exploitation and
optimization studies on over 1000 wells to
date, including both pressure and rate transient
analysis.

How do we use flowing pressures in decline analysis? Since production

rates vary proportionally with the flowing pressure drop pi pwf
(pseudo-pressure for gas), dividing the rates by the associated drop in
flowing pressure is a very effective method for normalizing production
data. If we plot the normalized rate against a special time function
called material balance time, we end up with a very powerful analysis
technique - constant rate and constant pressure production can be
made to appear the same.

Material Balance - Time

Graph B

Graph A

Q=Q1=Cumulative
Production at a given time
tc1

On Graph A, at time (t1) the cumulative production is Q1.

The rate, (q) at that time is used in Graph B and is assumed to be constant.
The time taken in Graph B for the well to flow at the constant rate to get the same cumulative production
as in Graph A, is tc1, or material balance time.
Graphs C and D below demonstrate the same principle, using a different time (t2) and the corresponding
material balance time tc2.

Graph C

Graph D

Actual Rate Decline

Constant Rate

1. Gas wells usually do not flow at a constant bottomhole pressure

throughout their lives
To illustrate this point, consider an exploratory well in a small gas field.
Suppose the well is produced at low back pressure, directly to inlet
facilities for a year. At this time, a development program is put in place,
whereby 10 additional wells per year are brought on stream, in a large,
neighboring gas field, and the surface gathering system is expanded. As
the new wells are brought on, the surface pressures begin steadily
increasing, due to line pack. Through several years of this type of
production, the original well may exhibit a false decline in production
rate, which is not representative of reservoir depletion, but rather is
purely a result of surface conditions. An exponential decline analysis of
the production data may significantly underpredict gas reserves.
However, by using flowing pressures in the analysis, a more accurate
estimate of gas-in-place is obtained.

Q=Q1

Material-Balance-Time

Actual Time

Avoid the Pitfalls of Conventional Decline Analysis

Taking production decline analysis to the next level can help to avoid
being misled by your data and short-changing your reserves. There are
several factors that can lead to problems when using basic decline
analysis techniques:

tc1 = Q1
q1

t1

Q
tc = q

Constant Rate

Actual Rate Decline

q1

q2

t2

tc2 =

Q=Q2

Q=Q2=Cumulative
Production at a given time

Actual Time

Q2
q2

tc2
Material-Balance-Time

Conceptually, material balance time is described as the amount of

time it would take to produce the current cumulative production at
the current rate. It is defined as cumulative production divided by
flow rate.
2. Gas compressibility is a very strong function of reservoir pressure
This is a phenomenon that is usually ignored during basic decline
analyses, but can have a dramatic effect on estimated gas-in-place,
particularly in high-pressure reservoirs. Compressibility, c (not to be
confused with compressibility factor z) is a fluid property which
describes the amount of volume of a fluid that can be moved with a
given change in pressure. It is a very important consideration when
dealing with reserves, because it describes the energy in the gas that
allows it to be produced from the reservoir. Since gas compressibility

ction Data - Advanced Decline Analysis

increases as the pressure decreases
, there are increasing amounts
of reservoir energy (per unit volume of gas) available as the reservoir is
depleted. Because decline analyses are often performed early in the life
of a well, when the reservoir pressure is high, the gas compressibility
effect is not evident and is not accounted for. Once the reservoir is
nearing depletion, the production data begins to flatten out, and often
far surpasses the original estimated ultimate recovery.

flow regimes. Plotting the data on typecurves clearly shows the

distinctions between flow regimes. In addition, these typecurves can be
used to provide estimates of transient parameters, such as permeability
and skin. Basic decline analysis focuses on recoverable reserves, while
advanced decline analysis can also determine original gas-in-place.

Advanced decline analysis accounts for changing compressibility with

reservoir pressure using a concept called drawdown pseudo-time. In
order to plot the pseudo-time function, the reservoir pressure at any
time must be known. Thus, the gas-in-place has to be estimated, prior
to calculating pseudo-time. Because of this complication, the usage of
pseudo-time becomes an iterative process. The GIP is estimated and
the analysis is performed to yield a new GIP estimate, which is then
substituted back into the pseudo-time function so that the process can
be repeated.

Most reservoir engineers are familiar with the P/Z versus cumulative
production plot. This is the well known classic material balance
analysis procedure for gas wells. The P/Z plot is a helpful complement
to rate versus cumulative production analysis because it yields the gasin-place, not just the recoverable reserves. The two are often used in
conjunction to yield single well or pool recovery factors. An important
restriction of the P/Z plot, however, is that it requires that you know
the average reservoir pressure at a given cumulative production.
Reservoir pressure data is usually obtained from measured shut-in
pressures, and no guarantee is provided as to whether these pressures
are stabilized. Thus, P/Z plots can often produce ambiguous results
due to poor data quality. In addition, because the collection of reservoir
pressure data requires wells to be shut-in, the data tends to be sparse,
since wells are flowing most of the time.

3. Tight gas wells often take months, even years to reach Boundary
Dominated Flow
The analysis of production data from tight (low permeability) gas wells
is very often done incorrectly, usually leading to overly pessimistic gasin-place estimates. The trick with tight gas production data is not in
identifying a trend, but in knowing whether that trend is indicative of
transient or boundary dominated flow, (or perhaps transitional).
This is not an issue for high permeability wells, because their production
enters the boundary dominated regime so quickly that a transient trend
in the data is usually not observable. However, if boundary dominated
flow has not yet been reached, regardless of how long a well has been on
production, it is impossible to predict reserves with any accuracy. The
best that can be done is to estimate a lower-bound value of reserves.
One way of determining whether or not a trend in production data is
indicative of boundary dominated flow, is to use advanced decline
analysis techniques. Production (along with flowing pressure) data can
be more accurately categorized into transient or boundary dominated

Estimate Gas-in-Place Using Only Production Data

It turns out that by using advanced decline analysis techniques, a

material balance can be performed without requiring any shut-in
pressures. The Flowing Material Balance, similar to the classic P/Z
plot, is based on flowing production data, rather than shut-in data. It
combines concepts from rate versus cumulative production analysis
with classic material balance analysis. It incorporates the concepts of
material balance time and pseudo-time to account for variable rate and
pressure history, and changing gas compressibility with reservoir pressure.
With the advent of real-time data acquisition and dissemination
systems (such as SCADA), accurate and complete flowing pressure data
sets are becoming commonplace. Thus, very little additional expense
is required to implement the advanced production decline techniques
that use flowing pressures. However,
the potential upside to using these
procedures to estimate reserves,
characterize reservoirs and identifiy
infill potential, is tremendous.

Feketes most recent software

release, F.A.S.T. RTATM,
integrates and enhances the
Arps, Fetkovich, Blasingame,
Agarwal-Gardner and NPI
methods of analyzing
production data to obtain
recoverable reserves and gasin-place. For more
information, contact Fekete at
403-213-4200 or 1-800625-2488.

Fekete Associates On the Road...

Fekete engineers travel around the world to visit clients.
Destinations last year in the United States included Colorado,
Oklahoma, Texas, and Utah. Internationally, we visited
China, Australia, and Pakistan. The purpose of these visits is
to complete project work, demonstrate software, and provide
technical support to our users.
A lunch and learn session in your office is a great way to train staff on how to use
our software, achieve the most benefit from the results, and learn tips on how to be
accurate and time efficient.
We also spend time one-onone with clients who are
working on challenging
projects with the software.
Its a great opportunity to
voice ideas on new
development or compare
strategies for getting the most
accurate results.
With the release of our most
recent software innovation,
Dave Lillico and Family on the beach in Australia
F.A.S.T. RTA TM (Rate
Transient Analysis), we will
likely be visiting near you. If you would
like to discuss any of our software, get
technical support, or schedule a visit,
please contact Kevin Dunn at
403-213-4200 or 1-800-625-2488.

Is there a technical
issue you would like us
to address in our
newsletter?
Comments, requests, or
questions about the content
of this newsletter can be
directed to [email protected]

Watch for the F.A.S.T.

VirtuWellTM
upgrade
scheduled for
release February
2002. The following upgrades will be
included:
Mechanistic Model
(new multiphase flow correlation)
Complex Wellbore Modeling
Simple Reservoir Model (kh, s, etc.)
Erosional Velocity
Chokes
Static Liquid Levels
Wellbore Schematic
Multiple Oil Property Correlations
New Interface

UPCOMING COURSES
UPCOMING
COURSES
Courses
are available
for most of
Courses are available for most
Feketes software packages. For up to
of Feketes software packages.
date
schedule
visit our
For up
to dateinformation,
schedule informawebpage
www.fekete.com
tion, visitat:our
webpge at:
To
register,
contact
www.fekete.com [email protected]
Tocall
register,
contact
Erin Mason
or
our course
coordinator
at 403at [email protected],
403-213213-4200
or 1-800-625-2488.

Marty Santo and Ralph McNeil on the

Great Wall of China

4200 or 1-800-625-2488.

We would also like to wish our best to

Sarah Williams, lead development
engineer for F.A.S.T. VirtuWellTM.
Sarah is on maternity leave as of
January this year. Rob Henderson,
another engineer in Feketes production
optimization group, will be overseeing
F.A.S.T. VirtuWellTM for the next year.

FEKETE ASSOCIATES INC.

2000, 540 - 5th Avenue S.W.

Calgary, Alberta, Canada T2P 0M2
Ph: (403) 213-4200 Fax: (403) 213-4298
Toll-Free: 1-800-625-2488
E-Mail: [email protected]
We invite you to review further information about Fekete on our

web site: http://www.fekete.com