Note: Descriptions are shown in the official language in which they were submitted.
CA 02714318 2010-09-08
CONTROL LOGIC METHOD AND SYSTEM
FOR OPTIMIZING NATURAL GAS PRODUCTION
FIELD OF THE INVENTION
The present invention relates to methods and apparatus for optimizing
production
in natural gas wells, particularly in gas wells producing with fixed-velocity
lift systems.
The present invention further relates to flow control valves adaptable for use
in
conjunction with gas well production optimization methods and systems,
particularly
including flow control valves that operate with no atmospheric venting.
BACKGROUND OF THE INVENTION
Natural gas is commonly found in subsurface geological formations such as
deposits of granular material (e.g., sand or gravel), coal, shale, or porous
rock. Production
of natural gas from these types of formations typically involves drilling a
well a desired
depth into the formation, installing a casing in the wellbore (to keep the
well bore from
sloughing and collapsing), perforating the casing in the production zone
(i.e., the portion
of the well that penetrates the gas-bearing formation) so that gas can flow
into the casing,
and installing a string of tubing inside the casing down to the production
zone. Gas can
then be made to flow up to the surface through a production chamber, which may
be
either the tubing or the annulus between the tubing and the casing.
Formation liquids, including water, oil, and/or hydrocarbon condensates, are
generally present with natural gas in a subsurface reservoir. For reasons
explained in
greater detail hereinafter, these liquids must be lifted along with the gas.
In order for this
to happen, one of the following flow regimes must be present in the well:
Pressure-Induced Flow
In a pressure-induced flow regime, the formation pressure (i.e., the pressure
of the
fluids flowing into the well) is greater than the hydrostatic pressure from
the column
of fluids (gas and liquids) in the production chamber. In other words, the
formation
pressure is sufficient to lift the liquids along with the gas. Pressure-
induced flow
-1-
CA 02714318 2010-09-08
occurs in wells producing from reservoirs having a non-depleting pressure;
i.e.,
where the reservoir pressure is high enough that production from the reservoir
results
in no significant drop in formation pressure. This type of flow regime is
common in
reservoirs under water flood or having an active water drive providing
pressure
support. Conventional gas lift technology may be used to enhance flow in a
pressure-
induced flow regime by lightening the hydrostatic weight of total fluids in
the
production chamber.
Pressure-induced flow is most commonly associated with wells that are
primarily oil-
producing wells, and is rarely associated with primarily gas-producing wells.
Velocity-Induced Flow
This type of flow occurs with gas reservoirs having a depleting pressure, and
it is
common in most gas reservoirs and all solution-gas-driven oil reservoirs. The
present
invention is concerned with velocity-induced flow, a general explanation of
which
follows.
In order to optimize total volumes and rates of gas recovery from a gas
reservoir, the
bottomhole flowing pressure should be kept as low as possible. The
theoretically
ideal case would be to have a negative bottomhole flowing pressure so as to
facilitate
100% gas recovery from the reservoir, resulting in a final reservoir pressure
of zero.
When natural gas is flowing up a well, formation liquids will tend to be
entrained in
the gas stream, in the form of small droplets. As long as the gas is flowing
upward at
or above a critical velocity (or "VCR" -- the value of which depends on
various well-
specific factors), the droplets will be lifted along with the gas to the
wellhead, where
the gas-liquid mixture may be separated using well-known equipment and
methods.
In this situation, the gas velocity provides the means for lifting the
liquids; i.e., the
well is producing gas by velocity-induced flow.
The critical velocity VCR for a given wellbore will be dependent on a number
of
factors which may vary from one wellbore to the next, and which are subject to
change over the life of the wellbore. Such factors may include, but are not
-2-
CA 02714318 2010-09-08
necessarily limited to, reservoir pressure, flow line pressure, liquid
production rate,
liquid composition, gas composition, and wellbore design.
Formation pressures in virgin reservoirs of natural gas tend to be relatively
high.
Therefore, upon initial completion of a well, the gas will commonly rise
naturally to the
surface by velocity-induced flow provided that the characteristics of the
reservoir and the
wellbore are suitable to produce stable flow (meaning that the gas velocity at
all locations
in the production chamber remains equal to or greater than the critical
velocity, VCR -- in
other words, velocity-induced flow). Typically these wells will flow with
excess velocity
(i.e., significantly greater than VCR), resulting in friction between the
flowing fluids and
the production chamber.
As gas reserves are removed, the formation pressure drops continuously,
resulting
in reduced fluid velocities in the wellbore. Lower fluid velocity provides the
benefit of
reduced friction loading; however, it also diminishes the water-lifting
capability of the
wellbore. Once the gas velocity has become too low to lift liquids, the
liquids
accumulate in the wellbore, and the well is said to be "liquid loaded". This
accumulation
of liquids results in increased bottomhole flowing pressures and reduced gas
recoveries.
In this situation, continued gas production from the well requires the use of
mechanical
methods and apparatus in order to remove liquids from the wellbore and to
restore stable
flow.
In summary, gas wellbores are subject to two types of loading: (1) friction
loading and (2) liquid loading.
Friction loading is caused by fluid flowing up the production tubing at high
velocity, and results in restricted formation drawdown. Friction loading
typically will not
"kill" a well (i.e., completely halt the production of well fluids); however,
it can
significantly restrict production. The remedy or solution for friction loading
is to reduce
fluid velocities -- e.g., by reducing gas flow rates (thus reducing gas
velocity), or by
increasing production chamber size while maintaining flow rates (thus reducing
gas
velocity).
-3-
CA 02714318 2010-09-08
Liquid loading is caused by insufficient fluid velocity up the production
tubing.
Like friction loading, liquid loading results in restricted formation
drawdown. Liquid
loading will eventually result in the well being killed. Any time a wellbore
is "killed"
(i.e., its production of well fluids is stopped) due to excessive liquids,
considerable costs
must be incurred to correct the problem and restore production from the well.
As discussed above, liquid loading can be reduced or eliminated by increasing
fluid flow velocity to produce velocity-induced flow conditions. However,
increased
flow velocity also promotes increased friction loading. Current wellbore
design methods
typically provide excess flow velocity (i.e., significantly higher than VCR)
to eliminate
liquid loading and ensure that the well does not die. The resultant friction
loading is
accepted as the lesser of two evils.
In order to maintain long-term stable production rates from any gas well, the
first
priority is to ensure that the wellbore is substantially free from
accumulation of liquids.
If liquids production is substantial, any accumulation of liquids in the
wellbore must be
removed, either continuously or intermittently depending on the rate of
accumulation. In
cases where liquids are removed on an intermittent basis, the well would
alternate
between production and clean-out cycles, with the clean-out cycle beginning
when
accumulated liquids reach an undesirable level, and with gas production
necessarily being
stopped during the clean-out cycle.
In accordance with the preceding discussion, it can be appreciated that
production
optimization for a gas well requiring removal of liquids can be achieved by
keeping the
fluid velocity as close as practically possible to the well's critical
velocity, in order to
prevent accumulation of liquids while minimizing friction loading.
U.S. Patent No. 6,991,034 (and corresponding International Application No.
PCT/CA2004/000478) teach methods and apparatus for enhancing natural gas well
productivity by maintaining a velocity-induced flow regime, thus providing for
continuous removal of liquids from the well and preventing or mitigating
liquid loading
and friction loading of the well. In accordance with US 6,991,034, a
supplementary
pressurized gas is injected into a first (or injection) chamber of a gas well
as necessary to
-4-
CA 02714318 2010-09-08
keep the total upward gas flow rate in a second (or production) chamber of the
well at or
above a minimum flow rate needed to lift liquids within the upward gas flow. A
cased
well having a string of tubing may be considered as having two chambers
(namely, the
bore of the tubing, and the annulus between the tubing and the casing), and
either of these
chambers can serve as the first (or injection) chamber, with the other serving
as the
second (or production) chamber.
The invention of US 6,991,034 provides for a gas injection pipeline, for
injecting
the supplemental gas into the injection chamber, and further provides a
throttling valve
(also referred to as a "choke") for controlling the rate of gas injection,
and, more
specifically, for maintaining a gas injection rate sufficient to keep the gas
flowing up the
production chamber at or above a set point established with reference to a
critical flow
rate. Strictly speaking, the critical flow rate is a well-specific gas
velocity above which
liquids will not drop out of an upward-flowing gas stream (as previously
explained).
However, the critical flow rate for a given wellbore may also be expressed in
terms of
volumetric flow based on the critical gas velocity and the cross-sectional
area of the
production chamber.
As explained in US 6,991,034, the critical flow rate for a particular well may
be
determined using methods or formulae well known to those skilled in the art. A
"set
point" (i.e., minimum rate of total gas flow in the production chamber) is
then selected,
for purposes of controlling the operation of the choke. The set point may
correspond to
the critical flow rate, but more typically will correspond to a value higher
than the critical
flow rate, in order to provide a margin of safety. Once the well has been
brought into
production, an actual total gas flow rate in the production chamber is
measured. If the
measured total gas flow rate (without gas injection) is at or above the set
point, the choke
will remain closed, and no gas will be injected into the well. However, if the
measured
total gas flow rate is below the set point, the choke will be opened so that
gas is injected
into the injection chamber at a sufficient rate to raise the total gas flow
rate in the
production chamber to a level at or above the set point.
-5-
CA 02714318 2010-09-08
Gas productions systems as taught in US 6,991,034 may be referred to as fixed-
velocity production systems.
U.S. Patent No. 7,275,599 (and corresponding International Application No.
PCT/CA2004/001567) teach methods and apparatus whereby the intake pipeline
running
between the production chamber of a natural gas well and the suction inlet of
an
associated wellhead compressor is completely enclosed within a vapour-tight
jacket of
natural gas under positive pressure (i.e., greater than atmospheric pressure).
Being
enclosed inside this "positive-pressure jacket", the intake pipeline is
"blanketed" with
positive-pressure gas and therefore is not exposed to the atmosphere at any
point. This
allows gas to be drawn into the compressor through the intake pipeline under a
negative
pressure (i.e., lower than atmospheric pressure), without risk of air entering
the intake
pipeline should a leak occur in the pipeline. If such a leak occurs, there
would merely be
a harmless transfer of gas from the positive pressure jacket into the intake
pipeline.
Should a leak develop in the positive pressure jacket, gas therefrom will
escape into the
atmosphere, and entry of air into the positive pressure jacket will be
impossible. As
taught in US 7,275,599, these teachings can be readily adapted for use in
conjunction
with wells producing gas under velocity-induced flow conditions in accordance
with
methods taught in US 6,991,034.
Extensive scientific research has developed a number of flow correlations that
predict downhole velocities in flowing wellbores. The oil and gas industry
relies on these
correlations to predict critical flow rates, and attempts to design production
tubing strings
such that fluid velocities will equal or exceed predicted critical velocities.
However,
experience with wellbore modeling indicates that some wells are capable of
lifting liquids
at velocities considerably below the predicted critical velocity, and some
wells can
become liquid loaded despite producing at fluid velocities well above the
predicted
critical velocity.
The ideal solution for optimizing a wellbore producing with a fixed-velocity
lift
system is to determine whether it requires continuous liquids removal, or
whether it
would be more optimally produced at velocities below critical with reduced
friction
-6-
CA 02714318 2010-09-08
loading, accompanied by intermittent removal of liquids. In other words, a
well that
liquid-loads over a period of (for example) ten days would benefit from
intermittent
clean-outs, while a well that loads over a period of one hour would require
continuous
liquids removal.
For the foregoing reasons, there is a need for systems and methods for:
= automatically determining whether continuous or intermittent clean-out of
liquids
is the optimal production mode for a given wellbore, having regard to wellbore
and formation properties;
= automatically determining the optimum duration of production and clean-out
cycles for wells utilizing an intermittent clean-out system;
= automatically determining the optimal production chamber (e.g., either the
tubing
or the casing annulus), and switching fluid flow in the wellbore accordingly;
and
= automatically determining actual critical velocities for producing
wellbores, and
for maintaining the set point substantially equal to the actual critical
velocities for
wellbores utilizing a continuous clean-out system.
Any system or method directed to the foregoing needs will necessarily entail
use
of flow control devices. Control valves of various types are commonly used to
control
the flow of both liquid and gaseous fluids. Flow control may be achieved using
a control
valve in combination with a controller (i.e., a device incorporating a
processor and a
memory, such as but not limited to a pneumatic controller or a programmable
logic
controller (PLC)) that compare one or more flow variables (such as but not
limited to
flow rate, pressure, and temperature) against pre-established values (or "set
points"). In
response to corresponding signals from the controller, the control valve
either opens
(partially or fully) or closes as necessary to maintain the flow variable(s)
in question at
the appropriate set point(s).
As used in this patent document, the term "control valve" may be understood as
referring to either a discrete control valve or to a control valve assembly
that incorporates
a control valve, according to the context. A typical conventional control
valve assembly
-7-
CA 02714318 2010-09-08
(such as, for example, the Fisher D (globe-style) or DA (angle-style) valve)
includes a
valve body, internal valve trim ("valve trim" being a term readily understood
by persons
skilled in the art), and a valve actuator. A conventional control valve
assembly will
commonly be provided in conjunction with additional accessories such as (but
not limited
to) positioners and proportional controllers. These accessories provide a
means to enter a
control set point for the control valve. Each configuration for a conventional
control
valve typically provides a relatively narrow capacity range for a given set of
process
conditions. Due to this narrow capacity range and the inherent complexity of
the control
valve, maintenance and design must be done by professional instrumentation
personnel.
As a result, conventional control valves are relatively expensive to purchase
and
maintain.
Most if not all control valves commonly used in the natural gas industry are
pneumatically driven, and their operation typically results in the venting of
methane gas
to the atmosphere. In the past, this undesirable operational characteristic
was considered
tolerable in view of the reliability of such pneumatically-actuated control
valves.
However, with the increasing focus on reducing greenhouse gas emissions and
improving
system efficiencies, there is an increasing incentive to find environmentally-
friendly
alternative methods and apparatus for controlling fluid flow and pressure.
Electric actuators can be used to eliminate the venting of gas. However,
electric
actuators are comparatively expensive and have significant electrical power
requirements,
with correspondingly high operating costs.
For the foregoing reasons, it is desirable to have a comparatively simple and
inexpensive control valve that can control fluid flow effectively under a
broad range of
process conditions with minimal power consumption and absolutely no external
venting
of gas. Such a control valve would ideally be serviceable by any ordinarily-
skilled field
personnel using comparatively inexpensive non-precision parts.
One known type of non-venting control valve is a bladder-type valve such as
the
"Sur-Flo"TM control valve manufactured by Sur-Flo Meters & Controls Ltd., of
Calgary,
Alberta, Canada. A typical bladder-type valve of the Sur-F1oTM type has a
valve core
-8-
CA 02714318 2010-09-08
comprising a pair of frustoconical sections configured much like common pails,
each
having a solid base at its small-diameter end and with its large-diameter end
being open,
but with its conical sidewall having a plurality of perforations. The two
frustoconical
sections are coaxially arranged inside a generally cylindrical valve housing,
with their
bases in close juxtaposition. A generally cylindrical elastomer sleeve (or
"bladder") is
disposed within the valve housing, completely encircling the frustoconical
sections, and a
pressure port is provided through the cylindrical sidewall of the valve
housing.
When this valve assembly is installed in a fluid flow line, and when there is
no
external pressure acting on the bladder through the pressure port, fluid can
flow freely
into the first frustoconical section and out through that section's sidewall
perforations
into the space between the frustoconical sections and the bladder, and then
into the
second frustoconical section through its sidewall perforations. However, if
sufficient
external fluid pressure is applied against the bladder via the pressure port
(such as from a
"volume bottle" or "expansion bottle" of well-known type, or from another
pressure
source), the bladder will contract against the frustoconical sections, sealing
off their
sidewall perforations, and thus preventing fluid flow through the valve
assembly. At
lower external fluid pressures, the bladder will restrict but not completely
prevent fluid
flow through the valve. Accordingly, flow through the valve can be
incrementally
controlled across the range between the fully-closed and fully-open position
by varying
the fluid pressure applied against the bladder via the pressure port, with the
degree of
flow restriction being roughly proportional to the external pressure acting
against the
bladder.
The bladder-type control valve is a comparatively simple non-venting valve
that
provides a wide range of flow control options under any process conditions.
This valve is
considerably less expensive than more complex control valves commonly in use,
and it is
readily serviceable by ordinarily skilled field personnel. This type of valve
has proven
durability and is very commonly used as a fixed-set-point back-pressure valve.
What is needed is an inexpensive adaptation of the bladder-type control valve
that
converts the simple fixed-set-point control valve to a variable-set-point
control valve,
-9-
CA 02714318 2010-09-08
thereby providing control of both flow and pressure. Such operational
capabilities will be
particularly beneficial in control valves used in conjunction with methods and
apparatus
for optimizing production in natural gas wells, but will be beneficial for
purposes of other
control valve applications as well.
BRIEF SUMMARY OF THE INVENTION
Control Logic Method and System
In general terms, the present invention teaches methods, systems and apparatus
for optimizing natural gas production in response to variations in one or more
selected
wellbore-specific parameters. Wellsite intelligence technology is used to
monitor the
liquid and gas production from the wellbore under friction-loaded conditions.
Wellsite
technology suitable for purposes of methods and systems in accordance with the
present
invention may take the form of one or more programmable logic controllers
(PLCs) or
other suitable programmable data-processing devices. For convenience, the term
PLC
will be used in this patent document as a general reference to wellsite
intelligence
technology, with it being understood that the scope of such references is not
restricted to
programmable logic controllers per se.
Using baseline data gathered during production tests, the PLC determines and
initiates the appropriate operating mode for the wellbore to optimize a
selected
production criterion (such as maximum gas production rate or maximum cash
flow), to
suit measured wellbore-specific parameters. The operating mode thus determined
will be
either a continuous clean-out mode, in which gas is continuously injected into
the
wellbore to prevent or minimize liquid loading, or an intermittent clean-out
mode, in
which liquid loading can be regulated by intermittent gas injection only. The
system can
be adapted to monitor wellbore performance on either a continuous or periodic
basis, and
to adjust the operational characteristics of the operating mode in effect at a
given time,
and/or to switch the operating mode, as may be dictated by variations in
measured
wellbore parameters.
-10-
CA 02714318 2010-09-08
In a first aspect, the present invention teaches a method for optimizing
production
from a natural gas well that is associated with a gas compressor, and which is
adapted for
injection of gas into a wellbore injection chamber to regulate wellbore
velocity. In one
embodiment, the method comprises the steps of-
= conducting a plurality of baseline production tests to gather selected well
productivity information under operational conditions corresponding to a
selected
set of test input parameters;
= storing the test input parameters and the corresponding well productivity
information in the memory of a programmable logic controller (PLC);
= by means of the PLC, identifying the set of test input parameters that
results in the
optimal well productivity characteristics; and
= operating the well using a first set of operational parameters points
corresponding
to the optimal input parameters determined in the "identifying" step.
The test input parameters may comprise (but are not limited to) the compressor
suction
pressure and the wellbore flow rate (i.e., gas flow rate up the wellbore). The
well
productivity information may comprise (but is not limited to) a rate of gas
production
from the well, or a rate of cash flow realized from the well.
In alternative embodiments, the method of the present invention may comprise
the
further steps of:
= conducting a plurality of supplementary production tests to gather selected
well
productivity information under operational conditions corresponding to a
selected
set of supplementary input parameters;
= storing the supplementary input parameters and the corresponding
supplementary
well productivity information in the memory of the PLC; and
= by means of the PLC, determining which supplementary input parameters
resulted
in the most favourable supplementary well productivity information, and:
o if the most favourable supplementary well productivity information
represents an improvement over well productivity achieved using the first
-11-
CA 02714318 2010-09-08
set of operational parameters, commence operating the well using the
supplementary input parameters; and
o if the most favourable supplementary well productivity information does
not represent an improvement over well productivity achieved using the
first set of operational parameters, revert to operating the well using the
first set of operational parameters.
In embodiments of the method, one or more control valve assemblies may be are
used to regulate wellbore velocity in accordance with the operative input
parameters.
The one or more control valve assemblies may be of any functionally suitable
design or
type, including control valve assemblies in accordance with embodiments of
control
valve assemblies described and illustrated in the patent document. In some
embodiments
of the method, the PLC may be adapted to control the operation of one or more
of the
control valves.
In a second aspect, the present invention teaches a system for regulating
multiple
separate fluid flows originating from a single fluid flow source. In one
embodiment, the
system comprises:
= an apparatus adapted to receive an inlet fluid flow, such that the apparatus
diverts
a first portion of the inlet fluid flow into a first downstream fluid flow,
and diverts
a second portion of the inlet fluid flow into a second downstream fluid flow;
= a first control valve associated with the first downstream fluid flow and
adapted to
regulate a first flow variable relative to a first set point range, with the
first flow
variable being the inlet fluid flow pressure;
= a second control valve associated with the second downstream fluid flow and
adapted to regulate a second flow variable relative to a second set point
range; and
= one or more additional downstream fluid flows, each having an associated
control
valve adapted to regulate an associated flow variable relative to an
associated set
point range.
In this embodiment, the first set point range is higher than the highest
downstream fluid
flow pressure, and the first control valve is the only valve regulating inlet
fluid pressure.
-12-
CA 02714318 2010-09-08
It will be apparent to persons skilled in the art that the system of the
invention can
be readily adapted to regulate more than separate fluid flows originating from
a single
fluid flow source, by using additional control valves as appropriate in
accordance with
the concepts described herein.
In some embodiments of the system, the set point ranges for all of the control
valves are wide enough to prevent valve chatter (i.e., rapid and repetitive
opening and
closing of a valve, typically due to the upper and lower set points being too
close to each
other). In some embodiments, the set point range of at least one control valve
is a fixed
set point range. In some embodiments, the set point range of at least one of
the control
valves is a variable set point range. In some embodiments, at least one
control valve may
have a fixed set point range, while at least one control valve has a variable
set point
range.
In one alternative embodiment, the system may further comprise a programmable
logic controller (PLC) having a memory storing the set point range of the
control valves.
In this embodiment, the PLC is adapted to regulate fluid flow through the
control valves
by carrying out the following steps:
= comparing, at selected time intervals, measured values of the flow variables
regulated by the control valves against the corresponding set point ranges;
and
= where a measured flow variable is outside its corresponding set point range
and
not trending toward the set point range, opening or closing the corresponding
control valve as appropriate to move the value of the flow variable toward its
set
point range.
The PLC may be adapted to open or close the control valves in pulsed fashion,
and to
compare measured flow variable values at selected time intervals.
The control valves used for the system of the invention may be of any
functionally suitable design or type, including control valve assemblies in
accordance
with embodiments of control valve assemblies described and illustrated in the
patent
document.
-13-
CA 02714318 2010-09-08
In accordance with another alternative embodiment of the system of the present
invention:
= the fluid flow source is a wellhead compressor associated with a natural gas
well,
with the gas well including a string of tubing disposed within the wellbore
and
defining an annulus surrounding the tubing;
= the fluid is natural gas from the well;
= the first downstream fluid flow is into an injection chamber selected from
the
tubing and the annulus, with the first downstream control valve regulating
compressor discharge pressure; and
= the second downstream fluid flow is into a sales flow line and the second
downstream control valve is regulated based on total production rate up a
production chamber selected from the tubing and casing other than the
injection
chamber.
In one variant of this embodiment, the compressor speed can be varied. In
another
variant embodiment, both the compressor throughput and discharge pressure have
fixed
set point ranges, such that the compressor speed varies directly with suction
pressure.
The system optionally may comprise a PLC having a memory that stores a
selected suction pressure set point range, with the PLC being adapted to
control the
compressor speed and thereby to regulate suction pressure by performing the
following
steps:
= comparing measured values of suction pressure against the stored suction
pressure
set point range;
= where the measured suction pressure is outside the stored suction pressure
set
point range and not approaching the set point range, increasing or decreasing
the
compressor speed by a selected incremental amount to move the value of the
suction pressure toward the stored suction pressure set point range; and
= repeating steps (a) and (b) at selected intervals until the measured suction
pressure
is within the stored suction pressure set point range.
-14-
CA 02714318 2010-09-08
The set point range of the suction pressure may be either fixed or variable.
In a third aspect, the present invention teaches a method for optimizing
production from a natural gas well associated with a gas compressor, where the
well
includes a string of tubing disposed within the wellbore and defining an
annulus
surrounding the tubing, with the method including the following steps:
= providing a system for regulating multiple separate fluid flows originating
from a
single fluid flow source, in accordance with any embodiment previously
described herein, where the system is adapted for selective diversion of the
second downstream gas flow into an injection chamber selected from the tubing
and the annulus;
= providing a PLC having a processor and a memory, with the PLC being
programmed to control the operation of the control valves and the compressor
speed, in response to input signals corresponding to selected flow variables;
= storing in the PLC memory a selected number of test data sets each
containing test
values for a selected set of flow variables;
= conducting a selected number of well production tests, during each of which
the
well operates under conditions corresponding to the flow variable values of a
selected test data set, with all well production tests being of the same
selected
duration;
= measuring the total volume of gas produced during each production test and
storing the measured gas volumes in the PLC memory with reference to their
corresponding test data sets;
= determining which production test produced the highest volume of gas, and
storing the corresponding test data set as the default operational parameters
for the
well in the PLC memory; and
= operating the well in accordance with the default operational parameters.
In variant embodiments, this method may comprise the additional steps of-
- 15-
CA 02714318 2010-09-08
= after a selected time interval, conducting a subsequent set of well
production tests
in accordance with the above-described "conducting" and "measuring" steps;
= determining which subsequent production test produced the highest volume of
gas, and identifying the test data set corresponding to that subsequent
production
test;
= if the test data set identified in the above-described "determining" step
does not
correspond to the default parameters stored in the PLC memory, establishing
the
test data set identified in the "determining" step as the new default
parameters and
store same in the PLC memory; and
= operating the well in accordance with the new default parameters.
In another method for optimizing production from a natural gas well associated
with a gas compressor, where the well includes a string of tubing disposed
within the
wellbore and defining an annulus surrounding the tubing, the method includes
the
following steps:
= providing a system for regulating multiple separate fluid flows originating
from a
single fluid flow source, in accordance with any embodiment previously
described herein, where the system is adapted for selective diversion of the
second downstream gas flow into an injection chamber selected from the tubing
and the annulus;
= providing a PLC having a processor and a memory, with the PLC being
programmed to control the operation of one or more of the control valves each
in
response to input signals from a sensor measuring a selected flow variable;
= storing, in the PLC memory, initial set point ranges for tubing flow rate,
suction
pressure, and discharge pressure;
= conducting an initial well production test for a selected test duration,
during
which the well operates under conditions corresponding to said initial set
point
ranges;
-16-
CA 02714318 2010-09-08
= logging selected well productivity data during the initial well production
test, and
storing said data in the PLC memory; and
= storing the initial set point ranges in the PLC memory as the default
optimum set
points.
In one alternative embodiment, this method may comprise the further steps of-
= after a selected time interval, adjusting the suction pressure set point
range by a
selected amount, while maintaining the initial tubing rate and discharge
pressure
set point ranges;
= conducting a supplemental well production test for test duration
corresponding to
the initial test duration;
= logging the selected well productivity data during the supplemental well
production test, and storing said data in the PLC memory; and
= comparing the logged well productivity data for the initial and supplemental
production tests; and
= if the supplemental well productivity data is more favourable than the
initial well
productivity data, establishing the supplemental set point ranges as the new
default parameters and storing same in the PLC memory.
In this embodiment, the adjustment to the suction pressure set point range may
correspond to a suction pressure reduction, or alternatively may correspond to
a suction
pressure increase.
Control Valve Assemblies
In preferred embodiments, the control logic methods and systems of the present
invention incorporate the use of a control valve assembly that enables the use
of variable
set points, in accordance with the present teachings. In one embodiment, the
control
valve assembly comprises:
(a) a bladder-type valve having a valve core, a valve bladder surrounding the
valve core, a fluid inlet, a fluid outlet, and a pressure port;
-17-
CA 02714318 2010-09-08
(b) a pressure source (such as a volume bottle, in a preferred embodiment)
connected to the valve's pressure port;
(c) an upstream bypass line connecting the fluid inlet and the pressure
source;
(d) an upstream solenoid valve operable to regulate fluid flow through the
upstream bypass line;
(e) a downstream bypass line connecting the fluid outlet and the pressure
source; and
(f) a downstream solenoid valve operable to regulate fluid flow through the
downstream bypass line.
Solenoid valves are well known in the art, and are commonly referred to simply
as solenoids. For convenience and simplicity, the term "solenoid" will be used
throughout this patent document to denote a solenoid valve.
The assembly described above allows fluid flow through the control valve at a
rate Q which will be approximately proportional to the pressure differential
Ap = (Pu -
PB), where Pu is the gas pressure upstream of the valve and PB is the gas
pressure acting
on the bladder. In other words, if the bladder pressure PB is equal to the
upstream
pressure Pu (i.e., PB = PU), the valve is locked closed, and flow through the
valve is zero.
When the bladder pressure PB is reduced to the downstream pressure PD (i.e.,
PB = PD),
the valve is wide open, resulting in maximum flow (QMAX) through the valve.
For
bladder pressure PB between PU and PD, flow through the valve will be
correspondingly
proportionate, relative to QMAX. For example, if PB = 0.5 (PU - PD), the flow
rate Q
through the valve will be approximately 50% of QMAx. Similarly, for PB = 0.35
(Pu -
PD), the flow rate Q through the valve will be approximately 35% of QMAx.
When the valve assembly is not in use, the upstream and downstream solenoids
will be closed. As a result, the bladder pressure PB will be constant,
resulting in a fixed
set point for the control valve. If the upstream pressure Pu increases above
this set point,
the valve will automatically open, allowing flow through the valve. If the
upstream
solenoid is held open, the upstream pressure Pu and the bladder pressure PB
will be in
equilibrium, such that the bladder will contract against the valve core and
close off fluid
-18-
CA 02714318 2010-09-08
flow through the valve. To commence flow through the valve, the downstream
solenoid
is opened, and the upstream solenoid is closed, thus reducing the bladder
pressure PB.
The fluid pressure acting against the bladder will thus be less than the
upstream pressure
Pu, so the bladder will resile away from the valve core and allow fluid flow
through the
valve. In preferred embodiments and preferred modes of use, fluid flow will be
controlled by pulsing the downstream solenoid to increase the flow rate Q and
pulsing the
upstream solenoid to decrease the flow rate.
In preferred embodiments of the control valve assembly, a controller such as a
programmable logic controller (PLC) is provided to control the operation of
the upstream
and downstream solenoids, in response to control signals from a control
sensor.
Depending on system requirements in specific applications, the control sensor
could
measure fluid pressure, fluid flow rate, fluid temperature, or other flow
variables. In one
embodiment, the control sensor is a motor RPM sensor, measuring the speed of a
motor
associated with a gas compressor. The location of a control sensor within the
control
valve assembly (e.g., whether it is upstream or downstream of the control
valve) may
vary depending on the particular flow variable being controlled, and depending
on
whether the variable is being sensed directly or indirectly.
In order to control a particular flow variable, a selected set point for the
flow
variable (for example, gas flow rate or gas pressure) - or, perhaps more
typically, an
allowable range between selected upper and lower values for the flow variable)
- is
stored in the PLC's memory. If the corresponding control sensor determines
that the
controlled flow variable is outside the allowable range, the PLC will pulse
the appropriate
solenoid to bring the variable back within the allowable range.
Although described and illustrated herein with reference to bladder-type
control
valves (such as the "Sur-Flo"TM valve previously mentioned), control valve
assemblies in
accordance with the present invention are not limited or restricted to the use
of those
particular types of valves.
Accordingly, in a third aspect, the present invention teaches a non-venting
control
valve assembly, for installation in a flow line carrying a fluid (gas or
liquid) under
-19-
CA 02714318 2010-09-08
pressure, and comprising: a valve having a fluid inlet; a fluid outlet; a
pressure port in
fluid communication with a pressure source; and a flow restriction element
exposed to
the pressure source via the pressure port. The flow restriction element is
adapted to fluid
decrease flow through the valve in response to increases in the pressure
source pressure,
and to increase flow through the valve in response to decreases in the
pressure source
pressure.
In a first embodiment, the control valve assembly also comprises an upstream
bypass line connecting the fluid inlet and the pressure source; an upstream
solenoid
operable to regulate fluid flow through the upstream bypass line; a downstream
bypass
line connecting the fluid outlet and the pressure source; and a downstream
solenoid
operable to regulate fluid flow through the downstream bypass line. When the
pressure
at the fluid inlet is greater than the pressure source pressure, opening the
upstream
solenoid will increase the pressure source pressure. When the pressure at the
fluid outlet
is less than the pressure source pressure, opening the downstream solenoid
will decrease
the pressure source pressure.
The control valve assembly may further comprise a programmable logic
controller (PLC) adapted to control the operation of the upstream and
downstream
solenoids. The PLC may be adapted to control the operation of the upstream and
downstream solenoids in response to data inputs from a pressure sensor
associated with
the fluid inlet. Alternatively or in addition, the PLC may be adapted to
control the
operation of the upstream and downstream solenoids in response to data inputs
from a
pressure sensor associated with the fluid outlet. Further, the PLC may be
adapted to
control the operation of the upstream and downstream solenoids in response to
data
inputs from a flow rate sensor associated with a fluid source upstream of the
fluid inlet.
In one variant embodiment of the control valve assembly, the valve may
comprise
a pair of frustoconical valve core sections, each having a perforated
frustoconical
sidewall, a solid end wall, and an opposing open end, with the solid end walls
of said
sections juxtaposed. In this variant embodiment, the flow restriction element
comprises a
generally cylindrical and deformable bladder surrounding the valve core, such
that:
-20-
CA 02714318 2010-09-08
= a sufficient increase in the pressure source pressure will deform the
bladder into
contact against the conical sidewalls of the valve core sections so as to
restrict
fluid flow through one or more perforations, thereby reducing the fluid flow
rate
through the valve;
= when the bladder is restricting fluid flow through one or more perforations,
but
not completely preventing fluid flow through the valve, a further increase in
the
pressure source pressure will further deform the bladder, thereby restricting
fluid
flow through one or more additional perforations, and thereby further reducing
the
fluid flow rate through the valve; and
= when the bladder is restricting fluid flow through one or more perforations,
a
decrease in the pressure source pressure will cause the bladder to resile away
from
the conical sidewalls of the valve core sections so as to allow fluid flow
through
one or more flow-restricted perforations, thereby increasing the fluid flow
rate
through the valve.
The flow restriction element of this embodiment may take the form of a bladder
similar
to the type used in "Sur-Flo"TM control valves, but is not limited to that
particular type of
flow restriction element.
In control valve assemblies in accordance with the present invention, one or
both
of the upstream and downstream solenoids optionally may be adapted for pulsed
operation, thereby facilitating incremental adjustments to the pressure source
pressure.
BRIEF DESCRIPTION OF THE DRAWINGS
Embodiments of the present invention will now be described with reference to
the
accompanying Figures, in which numerical references denote like parts, and in
which:
FIGURE 1 is a schematic diagram of the wellhead of a natural gas well
producing gas under velocity-induced flow conditions in accordance with prior
art methods taught in U.S. Patent No. 6,991,034.
-21-
CA 02714318 2010-09-08
FIGURE 2 is a schematic diagram of the wellhead of a natural gas well
producing gas under velocity-induced flow conditions and enclosed within a
positive pressure gas jacket in accordance with prior art methods taught in
U.S.
Patent No. 7,275,599.
FIGURE 3 is a graph of wellbore velocity versus time, schematically
illustrating the variable critical fluid flow velocity above which a gas well
will
be friction loaded and below which a gas well will be liquid loaded.
FIGURE 4 is a graph of wellbore velocity versus time, schematically
illustrating optimization of wellbore velocity, with the well operating in a
continuous clean-out mode in accordance with an embodiment of a gas
production optimization method of the present invention.
FIGURES 5A and 5B are schematic diagrams illustrating parameters for
determination of a wellbore's maximum liquid storage volume.
FIGURE 6A is a schematic diagram of a wellbore at the beginning of the
production cycle of the intermittent clean-out mode of an embodiment of a gas
production optimization method of the present invention.
FIGURE 6B is a schematic diagram of a wellbore as in FIG. 6A, shown at the
end of the production cycle.
FIGURE 7A is a schematic diagram of a wellbore at the beginning of the clean-
out cycle of the intermittent clean-out mode of an embodiment of a gas
production optimization method of the present invention.
FIGURE 7B is a schematic diagram of a wellbore as in FIG. 7A, shown at the
end of the clean-out cycle.
FIGURE 8 is a schematic diagram of a first embodiment of a bladder-type
control valve assembly in accordance with the present invention.
-22-
CA 02714318 2010-09-08
FIGURE 9 is a schematic diagram of a second embodiment of a bladder-type
control valve assembly in accordance with the present invention.
FIGURE 10 schematically illustrates the use of bladder-type control valves in
accordance with the present invention, to regulate the flow of sales gas and
injection gas in conjunction with embodiments of the method of the present
invention.
FIGURE 11 illustrates an example of a production test grid for use in
conjunction with gas production optimization methods in accordance with the
present invention.
DETAILED DESCRIPTION OF PREFERRED EMBODIMENTS
The teachings of the present invention may be best understood after first
reviewing methods and apparatus taught by the previously-mentioned U.S.
Patents No.
6,991,034 and No. 7,275,599 (the entire disclosures of which are incorporated
herein by
reference).
FIG. 1 schematically illustrates a natural gas well W penetrating a subsurface
formation F containing natural gas, with well W producing gas under velocity-
induced
flow conditions in accordance with one embodiment of the methods taught in
U.S. Patent
No. 6,991,034. Well W is lined with a casing 20 which has a number of
perforations
conceptually illustrated by short lines 22 within a production zone (generally
corresponding to the portion of the well penetrating the formation F). As
conceptually
indicated by arrows 24, formation fluids including gas, oil, and water flow
into the well
through perforations 22. A string of tubing 30 extends inside casing 20,
terminating at a
point within the production zone. The bottom end of tubing 30 is open, such
that fluids
entering the wellbore can freely enter tubing 30. An annulus 32 is formed
between
tubing 30 and casing 20. The upper end of tubing 30 runs into a surface
termination
apparatus or "wellhead" (not illustrated in detail), of which various types
are known in
the field of gas wells.
-23-
CA 02714318 2010-09-08
Tubing 30 and annulus 32 may be considered as separate chambers of well W. A
selected one of these chambers serves as the "production chamber" through
which gas is
lifted from the bottom of well W to the surface, while the other chamber
serves as the
"injection chamber", the purpose and function of which are explained in
greater detail
hereinafter. For purposes of the embodiment illustrated in FIG. 1, tubing 30
serves as the
production chamber, and annulus 32 serves as the injection chamber. In some
alternative
embodiments, tubing 30 serves as the injection chamber, and annulus 32 serves
as the
production chamber.
It should be noted that, to facilitate illustration and understanding of the
prior art
and the present invention, the Figures herein are not drawn to scale. The
diameter of
casing 20 is commonly in the range of 4.5 to 7 inches, and the diameter of
tubing 30 is
commonly in the range of 2.375 to 3.5 inches, while well W typically
penetrates
hundreds or thousands of feet into the ground. It should also be noted that
except where
indicated otherwise, the arrows in the Figures denote the direction of gas
flow within
various components of the apparatus.
In the well configuration shown in FIG. 1, tubing 30 serves as the production
chamber to carry gas from well W to a production pipeline 40, the downstream
end of
which is connected to the suction manifold 42S of a gas compressor 42. The
upstream
end of a discharge pipeline 41 connects at one end to the discharge manifold
42D of
compressor 42 and continues therefrom to a gas processing facility (not
shown). A gas
injection pipeline 16, for diverting production gas from production pipeline
40 for
injection into the injection chamber, is connected at one end to the discharge
pipeline 41
at a point X, and at its other end to the top of the injection chamber (i.e.,
annulus 32, in
FIG. 1). Also provided is a throttling valve (or "choke") 12, which is
operable to regulate
the flow of gas from production pipeline 40 into the injection chamber via
injection
pipeline 16.
Choke 12 may be of any suitable type. In a fairly simple embodiment of the
prior
art apparatus of FIG. 1, choke 12 may be of a manually-actuated type, which
may be
manually adjusted to achieve desired rates of gas injection, using trial-and-
error methods
-24-
CA 02714318 2010-09-08
as necessary or appropriate. Alternatively, choke 12 may be an automatic
choke; e.g., a
Kimray Model 2200 flow control valve. In alternative embodiments, a flow
controller
50 is provided for operating choke 12. Also provided is a flow meter 14
adapted to
measure the rate of total gas flow up the production chamber, and to
communicate that
information to flow controller 50. Flow controller 50 may be a pneumatic
controller of
any suitable type, such as a FisherTM Model 4194 differential pressure
controller.
Wellhead assemblies as schematically depicted in FIG. 1 will typically
incorporate a gas-liquid separator (not shown) in production pipeline 40
upstream of
compressor 42 for removing liquids present in the produced well fluids so that
only gas
flows to compressor 42.
In accordance with methods taught in U.S. Patent No. 6,991,034, a critical gas
flow rate is determined. The critical flow rate, which may be expressed in
terms of either
gas velocity or volumetric flow, is a parameter corresponding to the minimum
velocity
VCR that must be maintained by a gas stream flowing up the production chamber
(i.e.,
tubing 30, in FIG. 1) in order to carry formation liquids upward within the
gas stream
(i.e., by velocity-induced flow). This parameter is determined in accordance
with well-
established methods and formulae taking into account a variety of quantifiable
factors
relating to the well construction and the characteristics of formation from
which the well
is producing. A minimum total flow rate (or "set point") is then selected,
based on the
calculated critical flow rate, and flow controller 50 is set accordingly. The
selected set
point will preferably be somewhat higher than the calculated critical rate, in
order to
provide a reasonable margin of safety, but also preferably not significantly
higher than
the critical rate, in order to minimize friction loading in the production
chamber.
If the total flow rate measured by flow meter 14 is less than the set point,
flow
controller 50 will adjust choke 12 to increase the gas injection rate if and
as necessary to
increase the total flow rate to a level at or above the set point. If the
total flow rate is at or
above the set point, there may be no need to adjust choke 12. Flow controller
50 may be
adapted such that if the total gas flow is considerably higher than the set
point, flow
controller 50 will adjust choke 12 to reduce the gas injection rate, thus
minimizing the
-25-
CA 02714318 2010-09-08
amount of gas being recirculated to the well through injection, and maximizing
the
amount of gas available for processing and sale.
In the embodiment shown in FIG. 1, flow controller 50 has a computer with a
microprocessor 60 and a memory 62. Flow controller 50 also has a meter
communication
link 52 for receiving gas flow measurement data from flow meter 14. Meter
communication link 52 may comprise a wired or wireless electronic link, and
may
comprise a transducer. Flow controller 50 also has a choke control link 54,
for
communicating a control signal from computer 60 to a choke control means (not
shown)
which actuates choke 12 in accordance with the control signal from computer
60. Choke
control link 54 may comprise a mechanical linkage, and may comprise either a
wired or
wireless electronic link.
In operating the prior art apparatus of FIG. 1, the set point is stored in
memory 62.
Computer 60 receives a signal from flow meter 14 (via meter communication link
52)
corresponding to the measured total gas flow rate in the production chamber,
and, using
software programmed into Computer 60, compares this value against the set
point.
Computer 60 then calculates a minimum injection rate at which supplementary
gas must
be injected into the injection chamber, or to which the injection rate must be
increased in
order to keep the total flow rate at or above the set point. This calculation
takes into
account the current gas injection rate (which would be zero if no gas is being
injected at
the time). If the measured total gas flow is below the set point, computer 60
will convey a
control signal, via choke control link 54, to the choke control means, which
in turn will
adjust choke 12 to deliver injection gas, at the calculated minimum injection
rate, into
injection pipeline 16, and thence into the injection chamber of the well
(i.e., annulus 32,
in FIG. 1). If the measured total gas flow equals or exceeds the set point, no
adjustment
of choke 12 will be necessary, strictly speaking.
FIG. 2 conceptually illustrates a gas-blanketed natural gas wellhead in
accordance
with one embodiment of prior art apparatus taught by U.S. Patent No.
7,275,599, and
adapted to provide for gas injection into an injection chamber generally as in
the
apparatus of FIG. 1. The apparatus shown in FIG. 2 has numerous components
that
-26-
CA 02714318 2010-09-08
correspond to components of the apparatus of FIG. 1 (with like numerical
references).
Although not shown, the apparatus of FIG. 2 may incorporate a flow controller
50, choke
12, and associated components as in the apparatus of FIG. 1.
In the apparatus of FIG. 2, production pipeline 40 is divided into an upstream
section 40U that conveys produced well fluids under negative pressure from
well W's
production chamber (i.e., tubing 30 in FIG. 2) to a gas-liquid separator 70,
and a
downstream section 40D extending between separator 70 and suction manifold 42S
of
compressor 42. Similar to the apparatus of FIG. 1, a discharge pipeline 41
connects to
discharge manifold 42D of compressor 42 and continues therefrom to a gas
processing
facility (not shown). As schematically indicated, well fluids entering
separator 70
separate into a gas fraction that exits separator 70 through downstream
section 40D of
production pipeline 40 to compressor 42, and a liquids fraction 72 which,
being heavier
than the gas fraction of the well fluids, accumulates in a lower section of
separator 70.
Liquids 72 flow under negative pressure through a pump inlet line 78 to a pump
74,
which pumps liquids 72 (now under positive pressure) through a liquid return
line 76
which connects into discharge pipeline 41 at a point Z downstream of
compressor 42.
Alternatively, liquids 72 may be pumped to an on-site storage tank 80.
As illustrated in FIG. 2, upstream section 40U and downstream section 40D of
production pipeline 40, separator 70, and pump inlet line 78 are fully
enclosed by a
vapour-tight positive pressure jacket 150 that defines a continuous internal
chamber 152.
A gas recirculation pipeline 260 extends between, and is in fluid
communication with,
discharge pipeline 41 at point X located between compressor 42 and point Z,
and a
selected pressure connection point Y on positive pressure jacket 150, such as
between
compressor 42 and separator 70 as shown in FIG. 2. By means of recirculation
pipeline
260, a portion of the gas discharged from discharge manifold 42D of compressor
42 (at
positive pressure) may be diverted into positive pressure jacket 150, such
that upstream
section 40U and downstream section 40D of production pipeline 40, separator
70, and
pump inlet line 78 are entirely enclosed by a "blanket" of gas under positive
pressure.
Positive pressure jacket 150 thus enshrouds all components that contain
containing
combustible fluids under negative pressure with a blanket of gas under
positive pressure,
-27-
CA 02714318 2010-09-08
thereby preventing the entry of air in the event of a leak developing in any
component
containing combustible fluids under negative pressure. Blanketing gas may be
diverted
from internal chamber 152 for injection into a selected injection chamber of
well W via a
suitable injection pipeline represented by reference numeral 116 in FIG. 2,
for producing
well W under velocity-induced flow conditions as previously described in
connection
with FIG. 1. Alternatively, injection gas can be introduced into the injection
chamber via
a separate injection pipeline (not shown) connected into gas recirculation
pipeline 260.
Optimization of Production
FIG. 3 conceptually illustrates the previously-discussed principle that
production
optimization for a gas well requiring continuous removal of liquids can be
achieved by
keeping the upward fluid velocity in the production chamber as close as
practically
possible to the well's critical velocity, in order to prevent accumulation of
liquids while
minimizing friction loading. In FIG. 3 the line marked VCR (which may be
referred to as
the critical velocity curve) indicates the critical upward velocity for well
fluids flowing
up the production chamber of a gas well, at or above which liquid droplets
will be lifted
along with the gaseous portion of the well fluids; in other words, the
velocity at or above
which the well will be producing gas under velocity-induced flow conditions.
As conceptually illustrated in FIG. 3, and as noted earlier, the value of
critical
velocity VCR for a given wellbore is dependent upon a number of factors that
are subject
to change over the production life of the wellbore (such as flow line
pressure, reservoir
pressure, liquid production rate, liquid composition, gas composition, and
wellbore
design). Whenever the actual upward flow velocity in the production chamber
(also
referred to herein as the wellbore velocity VWB) is higher than the current
value of VCR
(i.e., above the critical velocity curve in FIG. 3), the well will be
producing under
velocity-induced flow conditions. The higher the wellbore velocity is relative
to VCR, the
greater the extent to which the wellbore will be subject to friction loading.
Whenever the
wellbore velocity is lower than the current value of VCR (i.e., below the
critical velocity
curve), the well will be prone to liquid loading.
-28-
CA 02714318 2010-09-08
FIG. 4 schematically illustrates how wellbore velocity VWB may be optimized
using methods and systems in accordance with the present invention; i.e., to
induce a
wellbore that maintains velocity-induced flow to minimize or prevent liquid
loading,
while also minimizing friction loading. Because wellbore-specific parameters
vary over
time, thus changing the wellbore's critical velocity, the system automatically
makes
upward or downward adjustments to the wellbore velocity VWB to keep it just
sufficient
to prevent liquid loading - in other words, effectively matching the flow rate
to the
critical velocity. This optimized production state is graphically illustrated
in FIG. 4, in
which the solid curve represents the critical velocity VCR, and the stepped
line indicates
the actual wellbore velocity VWB as periodically adjusted in accordance with
methods of
the present invention.
As may be generally understood from FIG. 4, production from the well in
accordance with methods of the present invention may be commenced in a
friction-
loaded state - i.e., at an initial wellbore velocity VWB having a value above
the critical
velocity curve in FIG. 3. The initial value for wellbore velocity VWB may be
determined
on a trial-and-error basis, or on the basis of a theoretical prediction of
critical velocity
VCR, but in either case erring on the high side to protect against starting
production with
the well in a liquid-loaded state.
At specified intervals, the system's PLC (or other suitable wellsite
intelligence
technology) gathers selected production data for the wellbore, such as liquid
flow rates,
gas flow rates, tubing pressure, and casing pressure, by means of suitable
sensing or
metering devices (which will be familiar to persons skilled in the art). The
PLC is
programmed to process this collected data to determine whether the wellbore
velocity
VWB needs to be increased or decreased to keep it close to the critical
velocity VCR and
thus prevent accumulation of liquids. This is accomplished without needing to
quantify
the critical velocity VCR, as the system determines the need to adjust
wellbore velocity
based on actual, real-time wellbore operational conditions. Any necessary
adjustments to
wellbore velocity will be automatically initiated by the PLC, which regulates
the
operation of gas flow control valves as described in detail later in this
specification.
-29-
CA 02714318 2010-09-08
Intermittent Clean-Out Mode
The PLC can also be programmed to run additional production tests at specified
intervals to determine whether the wellbore can be optimally produced with
intermittent
clean-out of liquids rather than on a continuous clean-out basis.
Wells accumulating (or "making") large quantities of water can load up within
minutes of wellbore velocities falling below the critical velocity VCR. For
such wells,
liquid loading can be alleviated or prevented by maintaining a wellbore
velocity VWB at
or above a critical velocity, as conceptually illustrated in FIG. 4. This
production mode
may be referred to as the continuous clean-out mode, in which all liquids are
removed
from the wellbore on a continuous basis.
In contrast, wells making comparatively low quantities of water can take days
to
load up (i.e., to become liquid loaded) after wellbore velocities have fallen
below the
critical velocity VCR. For such wells, it may be optimally efficient to
produce at wellbore
velocities below VCR, thus permitting a certain amount of liquid loading but
not enough
to kill the well, and removing accumulated liquids on an intermittent basis.
This
production mode may be referred to as the intermittent clean-out mode, in
which the
wellbore velocity VWB, although lower than the wellbore's critical velocity
VCR, is
sufficient to produce gas from the wellbore but will necessarily permit the
accumulation
and storage of liquids in the wellbore while gas is being produced.
In intermittent clean-out mode, particularly in wellbores that make minimal
quantities of water, it will generally be desirable to produce at a wellbore
velocity VWB
that is as low as possible to minimize friction loading. This may be done by
producing
up the production chamber that has the larger cross-sectional area, resulting
in a lower
wellbore velocity VWB for a given volumetric production rate. In typical
wellbores, this
will mean producing up annulus 32 rather than up tubing 30.
FIGS. 5A and 5B schematically illustrate the parameters for determining the
maximum volume of liquids that can accumulate in a wellbore without killing
the
wellbore. This maximum liquid storage volume, or LSVMAX, for a given wellbore
may
-30-
CA 02714318 2010-09-08
be defined as the maximum volume of liquid that can be stored in the wellbore
and
removed up the production chamber (typically tubing 30) by means of gas
injection into
the injection chamber (typically annulus 32). When LSVMAX is stored in tubing
30 and
annulus 32 as illustrated in FIG. 5A, with no differential pressure between
tubing 30 and
annulus 32, LSVMAX will come to a static height Hl above the bottom of the
wellbore.
As will be appreciated from FIG. 5A, static height Hl must not rise above
perforations 22
in casing 20, in order to prevent the well from being killed. Accordingly, the
static height
H1 of LSVMAX will be one variable in the determination of LSVMAX.
The value of LSVM x will also be dependent upon the capacity of the particular
compressor 42 associated with the wellbore. Stated another way, the
hydrostatic pressure
that would be produced when LSVMAX is completely contained within tubing 30
must be
less than the pressure that compressor 42 is capable of inducing by means of
gas injection
into annulus 32. This maximum hydrostatic pressure will be equal to the
vertical liquid
height H2 of LSVMAX if wholly contained within tubing 30 (as illustrated in
FIG. 5B)
multiplied by the specific gravity of the liquid.
An additional limiting factor with respect to the maximum liquid storage
volume
LSVMAx will be the reservoir pressure. If the hydrostatic pressure of LSVMAX
is greater
than the reservoir pressure, it will be impossible to clean out the
accumulated liquids
because the injection gas will simply flow into the subsurface formation
rather than
lifting the liquids. Therefore, the hydrostatic pressure of LSVMAX must not be
greater
than the lesser of the reservoir pressure and the compressor capacity.
Accordingly, the value of LSVMAX will vary from well to well, depending upon
wellbore dimensions, perforation height, compressor capacity, and reservoir
pressure. In
typical installations, the compressor will be capable of raising a column of
liquid in
tubing 30 under a hydrostatic head of about 700 kiloPascals. However, the
value of
LSVMAx for a particular well will typically be determined by calculation.
The intermittent clean-out mode is characterized by the alternating production
and
clean-out cycles, as follows:
-31-
CA 02714318 2010-09-08
Production cycle:
As schematically illustrated in FIGS. 6A and 6B, during the production cycle
the
well produces up the largest available production chamber (typically annulus
32), to
minimize friction loading. Due to the resultant low fluid velocities, however,
liquids
are not removed and gradually accumulate in the wellbore. Provided that the
wellbore is configured with tubing 30 submerged below perforations 22 in
casing 20,
as shown in FIG. 6B, the accumulated liquids will provide no loading on the
producing formation F. When the well's maximum liquid storage volume LSVMAX
has accumulated in the wellbore, the production cycle ends and the clean-out
cycle
begins.
Clean-out cycle:
During the clean-out cycle the well produces up the tubing 30, as
schematically
illustrated in Figs. 7A and 7B. The wellbore velocity will be above the
wellbore
critical velocity due to gas injection into the annulus 32, resulting in all
accumulated
liquids being removed. When liquids production ceases, the clean-out cycle
will
end, and a new production cycle will start.
The methods and systems of the present invention provide for testing a well to
assess whether optimal productivity will be realized operating in intermittent
clean-out
mode rather than continuous clean-out mode. This may be referred to as the
"mode test"
phase. After calculating the production cycle time (as described in detail
later herein),
the PLC will cause the well to produce in intermittent clean-out mode for a
predetermined time (for example, one week). Once this time has elapsed, the
PLC will
automatically change modes and produce in the continuous clean-out mode for
the same
predetermined time period. Upon completion of the "mode test" phase, the PLC
will
default into the mode that produced more gas during the test period (or that
produced the
greater cash flow, depending on the selected optimization criteria).
In order to easily and efficiently switch between production chambers in a gas-
blanketed wellbore (as described in U.S. Patent No. 7,275,599), a downhole
packer 26
-32-
CA 02714318 2010-09-08
with an integral three-way valve 28 (of any suitable type) may be used. The
operation of
such a three-way valve is schematically illustrated in Figs. 6A, 6B, 7A, and
7B. During
the production cycle of the intermittent clean-out mode as shown in FIGS. 6A
and 6B,
gas is produced up annulus 32 as well as up tubing 30, with three-way valve 28
being set
in a first position that allows gas to flow out of annulus 32 and into tubing
30 at an upper
region thereof. During the clean-out cycle as shown in FIGS. 7A and 7B, three-
way
valve 28 is set in a second position that allows supplementary or "recycle"
gas to be
injected into annulus 32 to remove accumulated liquids.
Monitoring real-time liquids production rates and suction pressures will
determine
the trigger point for ending the clean-out cycle, as well as providing a
benchmark gas-to-
liquid ratio for use in determining the duration of the production cycle. For
operations in
which the wellbore is not gas-blanketed, no downhole packer is required. In
such cases,
the three-way control valve does not need to be located downhole, and could be
located at
the surface.
In accordance with one embodiment of the method of the present invention,
production from a gas well is commenced in a friction-loaded state - i.e., at
a production
fluid velocity above the critical velocity curve in FIG. 3. This state may be
initiated on a
trial-and-error basis, or based on a theoretical prediction of the critical
velocity but erring
on the high side to protect against a liquid-loaded state. The PLC (or other
suitable
wellsite intelligence technologies) is used to gather production data for the
wellbore,
including liquid flow rates, gas flow rates, tubing pressure, and casing
pressure; this
production data is collected at regular intervals.
During this production testing, the following data are captured or determined
for
each blowcase cycle (i.e., each time accumulated liquids are removed from the
blowcase
of separator 70):
1. time since the last blowcase cycle (minutes);
2. liquid volume dumped during the current blowcase cycle (litres);
3. gas volume produced since the last blowcase cycle (in cubic meters x 103);
- 33 -
CA 02714318 2010-09-08
4. water-to-gas ratio (WGR) for the current blowcase cycle, in liters of water
per 103
cubic meters of gas (calculated using values from items 2 and 3 above); and
5. casing (i.e., annulus) pressure at the start of the current blowcase cycle.
Utilizing the calculated WGR and maximum liquid storage volume LSVMAX for
the wellbore, and while the wellbore is operating in the intermittent clean-
out mode, the
system's PLC calculates the duration of the next production cycle (i.e., how
long it will
take for LSVMAX to accumulate in the wellbore). This value is determined in
accordance
with the following formula:
Production Cycle Time = LSVMAX / (WGR x QAV)
where QAV is the average volumetric gas flow rate.
After production from the well is stabilized in a friction-loaded state, the
wellbore
velocity VWB is gradually reduced in order to effectively eliminate friction
loading. The
PLC does this by initiating incremental changes to VWB at predetermined time
intervals,
while continually comparing real-time wellbore data against the baseline data
in order to
detect the onset of liquid loading as VWB is reduced. Imminent liquid loading
will
typically be indicated by an upward trend in casing pressure and a downward
trend in
liquid production rates. Other "flags" may also be used to identify the
beginning or
approach of a liquid-loaded state. Upon sensing the onset of liquid loading,
the
optimization system of the present invention automatically increases the rate
of gas
injection to raise the wellbore velocity VWB by an increment sufficient to
return the
wellbore to a marginally friction-loaded state.
Because wellbore-specific parameters vary over time, thus changing the
wellbore's critical velocity, the system automatically makes upward or
downward
adjustments to the wellbore velocity VWB to keep it just sufficient to prevent
liquid
loading - in other words, effectively matching the flow rate to the changing
critical
velocity, based on real-time wellbore data. This optimized production state is
graphically
illustrated in FIG. 4, in which the solid curve represents the critical
velocity, and the
-34-
CA 02714318 2010-09-08
stepped line indicates the actual gas flow rate as periodically adjusted in
accordance with
the present invention.
Critical Velocity Determination
In accordance with the method and system of the present invention, the
critical
velocity for a producing gas well is automatically determined, and the set
point (i.e.,
upward fluid velocity in the production tubing) is automatically adjusted as
may be
necessary from time to time in response to changes in the critical velocity,
thereby
maintaining a set point substantially equal to the critical velocity, thus
preventing liquid
loading and minimizing friction loading in the wellbore.
Using the stabilized WGR and average gas production rate, the PLC can
calculate
an estimated time until the next blowcase dump cycle, in accordance with the
following
formula:
Time until next dump = Time of last dump -
[Volume per cycle / (WGR x QAV)1
When the wellbore velocity VWB falls below the critical velocity VCR, liquids
will
accumulate in the wellbore, resulting in an increase in the time between
blowcase dumps.
This increase in time between dumps will serve as the primary flag for
indicating the
onset of liquid loading.
A secondary flag for liquid loading will be an upward trend in the pressure in
annulus 32. This is a proven method of identifying wellbore loading commonly
used in
most wellsite optimization systems currently available.
Other "flags" may also be used to help identify the beginning or approach of a
liquid-loaded state. Upon sensing the onset of liquid loading, the
optimization system of
the present invention automatically increases the gas flow rate in the
production tubing
by an increment sufficient to return the wellbore to a marginally friction-
loaded state.
-35-
CA 02714318 2010-09-08
Gas Flow Regulation
Optimization of gas production in accordance with the methods described above
necessarily entails the use of flow control devices such as control valves to
regulate the
rate of gas injection into the well. FIG. 8 schematically illustrates a basic
embodiment of
a control valve assembly 100 in accordance with the present invention, and in
the specific
context of a natural gas flow line. In the embodiment of FIG. 8, the control
valve
assembly 100 includes a bladder-type control valve 120 a valve core 122
disposed within
a valve housing 125, with valve housing 125 having an intake port 127, an
outlet port
128, and a pressure port 126. Valve core 122 comprises an upstream
frustoconical section
122U and a downstream frustoconical section 122D, each having a solid base at
its small-
diameter end and with its large-diameter end being open, but with its conical
sidewall
having a plurality of perforations 123. The two frustoconical sections 122U
and 122D
are coaxially arranged inside a generally cylindrical valve housing, with
their bases in
close juxtaposition. A generally cylindrical bladder 124, made from of a
resilient,
deformable material, is disposed around surrounding valve core sections 122U
and 122D.
In FIG. 8, bladder 124 is shown contracted against the conical sidewalls of
frustoconical
sections 122U and 122D, in response to differential pressure acting on bladder
124,
thereby blocking flow through perforations 123, such that gas entering
upstream valve
core section 122U cannot flow into downstream valve core section 122D.
Control valve assembly 100 of FIG. 8 further comprises a fluid inlet line 130
in
fluid communication with intake port 127; a fluid outlet line 140 in fluid
communication
with outlet port 128; and a pressure port 126. A pressure source 150 (such as
a volume
bottle, in a preferred embodiment) is connected to pressure port 126. An
upstream
bypass line 132 connects fluid inlet line 130 to pressure source 150, with an
associated
upstream solenoid 134 operable to regulate fluid flow through upstream bypass
line 132.
A downstream bypass line 142 connects fluid outlet line 140 to pressure source
150, with
an associated downstream solenoid 144 operable to regulate fluid flow through
downstream bypass line 142.
In FIG. 8, PU denotes gas pressure upstream of control valve 120, PB denotes
gas
pressure acting on bladder 124 of control valve 120, and PD denotes gas
pressure
-36-
CA 02714318 2010-09-08
downstream of the control valve. Any time the system is not running, upstream
solenoid
134 will be open and downstream solenoid 144 will be closed, such that Pu will
be equal
to PB, and control valve 120 will therefore be closed as shown in FIG. 8, with
bladder
124 blocking flow through perforations 123 in valve core sections 122U and
122D. After
the system had been started, upstream solenoid 134 remains open, thus ensuring
no gas is
lost until it is desired to commence flow through valve 120.
Pulsing downstream solenoid 144 open causes PU to exceed PB, thus initiating a
flow of gas through control valve 120. The first time downstream solenoid 144
is pulsed
to initiate flow, upstream solenoid 134 is automatically closed. After this
point both
upstream solenoid 134 and downstream solenoid 144 remain closed unless they
are
pulsed open. Pulsing downstream solenoid 144 open will increase the gas flow
rate,
whereas pulsing upstream solenoid 134 open will decrease the flow rate. If the
valve
assembly goes down for any reason, upstream solenoid 134 will automatically
open,
restoring pressure equilibrium (i.e., PU = PB), such that gas flow through
control valve
120 is shut off.
FIG. 9 illustrates an embodiment of the flow control valve assembly using a
PLC
160 in conjunction with a pressure sensor 170 installed upstream of control
valve 120,
with PLC 160 being in electronic communication with pressure sensor 170 via a
pressure
sensor link 172. In addition, PLC 160 is in electronic communication with
upstream
solenoid 134 and downstream solenoid 144, via upstream and downstream solenoid
links
162 and 164 respectively, for controlling the operation of solenoids 134 and
144.
For purposes of flow control, this embodiment of control valve 120 may be
operated by the following steps, as controlled by PLC 160:
1. Close control valve 120 by energizing upstream solenoid 134 to equalize
upstream pressure PU and bladder pressure PB. This provides a known starting
point for the logic process and ensures a safe start-up. Any time upstream
solenoid 134 is open for an extended duration, control valve 120 is closed.
2. Open control valve 120 by pulsing downstream solenoid 144 open, such that
PB
drops below PU due to the flow of gas from volume bottle 150 through
-37-
CA 02714318 2010-09-08
downstream solenoid 144. This results in a pressure differential across
bladder
124, urging bladder 124 radially outward from valve core sections 122U and
122D and thereby allowing gas to flow through perforations 123. When
downstream solenoid 144 is initially pulsed, upstream solenoid 134 is
automatically de-energized. After this point, upstream solenoid 134 and
downstream solenoid 144 both remain closed unless they are pulsed open.
3. Normal operation of control valve 120 in flow control mode: After each
pulse of
a solenoid, PLC 160 compares the actual volumetric gas flow rate (based on
data
received from a suitable flow meter, not shown in FIG. 9) to a desired
operational
envelope (i.e., a specified range of acceptable flow rates) stored in the
memory of
PLC 160, whereupon:
- If the flow rate is below the desired envelope, downstream solenoid 144 is
pulsed, thus increasing the flow rate.
- If the flow rate is above the desired envelope, upstream solenoid 134 is
pulsed, thus decreasing the flow rate.
- If the flow rate is within the desired envelope, neither solenoid is pulsed.
For purposes of pressure control, the system illustrated in FIG. 9 may be
operated
by the following steps:
1. Both upstream solenoid 134 and downstream solenoid 144 are closed, thus
locking in a specified bladder pressure PB.
2. Normal operation of control valve 120 in pressure control mode: After each
pulse
of one of the solenoids, PLC 160 compares the actual pressure (as sensed by
pressure sensor 170) to a desired operational envelope (i.e., a specified
range of
acceptable flow pressures), whereupon:
- If the pressure is below the desired envelope, upstream solenoid 134 is
pulsed, thus increasing bladder pressure PB.
- If the pressure is above the desired envelope, downstream solenoid 144 is
pulsed, thus decreasing bladder pressure PB.
-38-
CA 02714318 2010-09-08
- If the pressure is within the desired envelope, neither solenoid is pulsed.
Control valve arrangements in accordance with the present invention offer a
number of significant advantages over known technologies, including the
following:
= The system is entirely vent-free, thus eliminating all emissions of gas
(e.g.,
methane) to atmosphere, and allowing all gas to be sent to sales;
= The system eliminates methane concentrations in separator / compressor
buildings, thus providing a safer working environment;
= Purchase costs for the system are considerably lower than alternative
control
valve assembly currently available;
= The control valve can accommodate a certain amount of solid matter;
= Valve noise is eliminated under all process conditions;
= Two control valves can work efficiently in the same fluid system. Valve
chatter
is eliminated due to gentle valve response; and
= Maintenance costs are lower than for known alternative control valve
arrangements, as the comparatively simple design allows the control valve to
be
easily serviced by most operators.
The block diagram of FIG. 10 schematically illustrates how bladder-type
control
valves as taught herein may be used to regulate two or more separate gas flows
originating from a single fluid flow source. This is illustrated in FIG. 10,
by way of non-
limiting example, in the specific context of a first flow of gas from a well
to a processing
and sales facility, and a second flow of gas intended for injection into a
selected injection
chamber of the well for purposes of producing or maintaining velocity-induced
flow
conditions as previously described herein.
In FIG. 10, process block 200 conceptually represents wellhead apparatus
comprising a compressor and other components (typically including a separator)
as may
be required for a given gas well. A gas supply flowline 210 (analogous to
production
pipeline 40 in FIG. 1 or upstream production pipeline 40U in FIG. 2) conveys
produced
-39-
CA 02714318 2010-09-08
fluids from the well to wellhead apparatus 200. A volumetric flow meter 220 is
provided
in association with supply flowline 210. Wellhead apparatus 200 processes the
well
fluids to produce a first flow of gas into a sales gas pipeline 230 downstream
of the
compressor and a second flow of gas into an injection pipeline 240 (analogous
to
injection pipeline 16 in FIG. 1 or injection pipeline 260 in FIG. 2).
A first control valve CV-1, provided in association with sales gas pipeline
230, is
generally in accordance with the embodiment shown in FIG. 8, with an upstream
solenoid 134-1, a downstream solenoid 144-1, and a volume bottle (or other
pressure
source) 150-1. First control valve CV-1 is in electronic communication with a
PLC 160
for regulating the operation of upstream and downstream solenoids 134-1 and
144-1 as
previously described in connection with FIG. 9. PLC 160 is also in electronic
communication with flow meter 220. Accordingly, first control valve CV-1 is
configured
to control the volumetric flow rate in supply flowline 210 by regulating the
flow of gas in
sales pipeline 230.
A similar second control valve CV-2 is provided in association with injection
pipeline 240, and has an upstream solenoid 134-2, a downstream solenoid 144-2,
and a
volume bottle 150-2, plus a pressure sensor 170 as in the embodiment of FIG.
9. Second
control valve CV-2 is in also electronic communication with PLC 160 (or, in
alternative
embodiments, with a separate PLC) for regulating the operation of upstream and
downstream solenoids 134-2 and 144-2. As in the embodiment of FIG. 9, PLC 160
is
also in electronic communication with pressure sensor 170. Accordingly, second
control
valve CV-2 is configured to control pressure.
As indicated above, first control valve CV-1 controls the flow rate in supply
flowline 210. First control valve CV-1 slowly begins to close if the tubing
rate is below a
selected lower tubing rate set point (or LTRSP) stored in PLC 160's memory. As
first
control valve CV-1 closes, more gas is recycled to the well's selected
injection chamber
via injection pipeline 240, thereby causing the tubing flow rate to increase.
First control
valve CV-1 slowly begins to open if the tubing rate is above a selected upper
tubing rate
set point (or UTRSP) stored in PLC 160's memory. As first control valve CV-1
opens,
-40-
CA 02714318 2010-09-08
less gas is recycled to the injection chamber, thus causing the tubing rate to
decrease.
First control valve CV-1 will have no action if the tubing rate is in the
operational
envelope between the UTRSP and the LTRSP.
As indicated previously, second control valve CV-2 controls upstream gas
pressure. Although second control valve CV-2 is illustrated as having an
upstream
solenoid 134-2 and a downstream solenoid 144-2, in alternative embodiments
second
control valve CV-2 can have a fixed set point with no solenoids. In that
scenario, the set
point will need to be set above the peak flow line pressure typically seen
during normal
operation. This will allow first control valve CV-1 to direct gas either to
sales or recycle.
If first control valve CV-2 has a fixed set point below the flowline pressure
it will be
impossible to send gas to sales.
For ultimate efficiency, however, second control valve CV-2 can incorporate
solenoids as shown in FIG. 10, to provide a variable pressure set point. In
this scenario,
the set point will be automatically be maintained at a fixed amount above the
flowline
pressure. As the flowline pressure changes, so will second control valve CV-
2's set point.
Comprehensive field trials using fixed-velocity gas production systems as
taught
in US 6,991,034 have demonstrated that a given wellbore will not necessarily
load up
with liquids and die if the LTRSP is significantly below the wellbore's
critical rate (i.e.,
critical velocity VCR). FIG. 3 illustrates how the critical rate is the point
at which both
friction loading and liquid loading are minimized. As a result, any time a
wellbore is
produced exactly at the critical rate, the wellbore is achieving maximum
drawdown and
achieving maximum gas production.
The second part of the optimization process is to determine the speed
(measured
in RPM) at which the compressor should run. The design of any compressor will
dictate
that at any given throughput and discharge pressure there will be a
corresponding suction
pressure. The tubing rate set point (or TRSP) defines the throughput and the
flowline
pressure defines the discharge pressure, so compressor speed will be directly
related to
suction pressure. By establishing a desired operational suction pressure set
point, the
production optimization system of the present invention will gradually speed
up the
-41-
CA 02714318 2010-09-08
compressor until this suction pressure set point is met (provided that the
compressor
motor is not overloaded).
Various energy saving routines can be added to reduce power consumption. For
example, the system can identify when the gas processing plant goes down and
the
system is in full recycle (i.e., with no gas going to sales). As well, the
compressor speed
can be reduced to minimize wasted power. It is also a good idea to protect the
system by
establishing a maximum compressor speed. This would be an RPM set point
slightly
above the typical operating RPM.
In accordance with production optimization methods of the present invention, a
fixed-velocity system will have the ability to maintain the tubing gas rate
within pre-set
upper and lower tubing rate set points (UTRSP and LTRSP). However, if the
UTRSP is
too high, excessive gas flow velocity in the wellbore's production chamber
will restrict
production due to friction loading. If the LTRSP is too low, liquid build-up
in the
wellbore will restrict production due to liquid loading. In this scenario, the
wellbore's
production will become very unstable, intermittently loading up and then
unloading itself
Comprehensive field trials using our fixed-velocity system have confirmed that
a
wellbore will not load up and die if the LTRSP is significantly below the
critical rate for
the subject wellbore. The RPM controls automatically adjust the compressor
speed to
bring the suction pressure to the set point. Reduced suction pressure will
increase tubing
velocities, removing liquids from the wellbore but causing increased friction
loading.
Optimum Set Point
A further aspect of the production optimization methods of the present
invention
is the ability to automatically determine the tubing flow rate and the
corresponding
suction pressure that will result in optimal production from a given wellbore.
This
optimal combination of tubing rate and suction pressure may be referred to as
the
optimum set point (or OSP).
A wellbore's OSP can be defined in different ways, depending on selected
criteria. The OSP would most commonly be defined to maximize gas production -
which
-42-
CA 02714318 2010-09-08
in most conditions will correspond to maximum cash flow from the wellbore. In
some
situations, however, maximum production might not equate to maximum cash flow.
For
example, power consumption and water production might not be directly
proportional to
the gas production rate, such that the incremental power cost to increase well
production
could be more than the incremental value of gas produced. In a different
scenario, the
water-to-gas ratio (WGR) in a producing gas well might increase
disproportionately with
increased gas production rates, thus making it advantageous to produce the
well at lower
rates. Accordingly, it may be desirable for some wellbores to define the OSP
to maximize
cash flow rather than to maximize production, and the criteria for
establishing the OSP
might vary over the life of the wellbore. It is also possible that the OSP for
some
wellbores would be based on criteria other than production or cash flow.
TABLE 1
Tubing Flow Suction Average Sales Average Cash
Rate set point Pressure Rate Flow
(m3 X 103 (m3 X 103
per day) (kPa) per day) ($ per day)
5 200
5 150
5 100
5 50
5 0
5 -50
4 200
4 150
4 100
4 50
4 0
4 -50
3 200
3 150
3 100
3 50
3 0
Implementation of optimization methods in accordance with the invention may be
facilitated by creating a production test grid for the wellbore, by conducting
a series of
-43-
CA 02714318 2010-09-08
baseline production tests using selected values for key operational
parameters. Table 1
above illustrates a sample production test grid which may be maintained in the
PLC
memory to record input data and well performance data for each production
test.
If it is desired to optimize production from the well on the basis of gas
production
rate, the production test grid may be defined using selected input values for
the following
operational parameters, which are programmed into the PLC:
= Tubing rate set point (TRSP) at start of test;
= TRSP increment;
= TRSP at end of test;
= Suction pressure - start;
= Suction pressure - end; and
= Test duration (hours).
The PLC totalizes the gas production during each test period. Back pressure is
kept constant for all tests to ensure that a true measurement of wellbore
productivity is
provided. After a desired number of these production tests have been
completed, with
the test results being used to populate Table 1 as appropriate, the
combination of TRSP
and suction pressure that resulted in the highest gas production during the
tests will
represent the optimum set point (OSP) for the wellbore. The PLC will save
these values
as the default operating parameters for the wellbore, and the system will
revert to
operating at these optimum conditions once the testing is complete.
FIG. 11 provides a pictorial representation of data used for purposes of a
production test grid as described above, with tubing rate being plotted
against compressor
suction pressure, with further reference to compressor speed (RPM). Each
circular data
point in FIG. 11 represents a particular combination of tubing rate and
suction pressure
inputs used in production testing.
If it is desired to optimize cash flow rather than gas production rate,
additional
inputs may be required for the production tests, such as:
= Water disposal cost (per cubic meter);
-44-
CA 02714318 2010-09-08
= Gas price (per cubic meter); and
= Power cost (per kilowatt-hour).
The PLC totalizes gas and water production as well as power consumption during
each
test period. Back pressure is kept constant for all tests to ensure that a
true indication of
wellbore productivity is provided. The PLC uses the measured gas and water
production
data and the corresponding cost and price information to determine a cash flow
indicator.
After a desired number of these production tests have been completed, with the
test
results being used to populate Table 1 as appropriate, the combination of TRSP
and
suction pressure that resulted in the highest cash flow during the tests will
represent the
optimum set point (OSP) for the wellbore. These values will be saved as the
default
operating parameters for the wellbore, and the system will revert to operating
at these
optimum conditions once the production testing is complete.
After test procedures have been performed as described above, the system will
have identified the parameters for optimum production with continuous removal
of
liquids. The optimization test grid will preferably be run on a periodic basis
throughout
the life of the well. In most cases, a very broad range of suction pressures
and tubing gas
rate values will be used the first time the test grid is run. Subsequent
optimization runs
will concentrate the test grid around the default optimum conditions
determined from
earlier tests. Based on the average time between liquid dumps, the system will
determine
whether the well should be tested for intermittent liquids removal. For
example, if the
time between dumps is 24 hours, there is a good chance that production
improvements
can be realized through intermittent liquid removal. In contrast, if the time
between
liquid dumps is, say, only 8 minutes, intermittent removal of liquids would
not be an
option, because this would almost certainly have the effect of killing the
well.
-45-
CA 02714318 2010-09-08
Determining Viability of Intermittent Clean-Out
To determine whether intermittent removal of liquids is viable for a given
wellbore, a new test grid is set up on the PLC, such as illustrated in Table 2
below:
TABLE 2
Average Cash
Sub-critical flow time Average Sales Rate Flow ($ per day)
(minutes) m3 x 103 per day)
Avg. dump time x.5
Avg. dump time x.6
Avg. dump time x.7
Avg. dump time x.8
Avg. dump time x.9
Avg. dump time x 1.0
Sub-critical flow is a flow rate in the wellbore that does not provide
sufficient
velocity to lift liquids. Accordingly, the sub-critical flow time in Table 2
is the time
during which the wellbore operates under sub-critical flow conditions. During
this time
liquids are slowly accumulated in the wellbore; however, friction loading is
eliminated.
Sub-critical flow can occur up the wellbore annulus alone or up the tubing and
the
annulus together.
If it is desired to assess the viability of intermittent clean-out for a given
wellbore
on the basis of optimizing gas production rate, the production test grid may
be defined
using selected input values for the following operational parameters:
= Initial sub-critical flow duration, as a multiple of dump time (using the
dump time
multiple helps eliminate excessive liquid accumulations);
= Sub-critical flow duration increment (multiple of dump time);
= Clean-out cycle duration (minutes);
= Suction pressure set point (as determined from the continuous clean-out test
grid);
and
= Test duration (number of cycles).
-46-
CA 02714318 2010-09-08
The PLC totalizes the gas production during each test period, and then divides
this total
by the test duration to calculate an average daily gas rate. Back pressure is
kept constant
for all tests to ensure that a true indication of wellbore productivity is
provided.
Production test results are used to populate Table 2 as appropriate. At the
completion of
each sub-critical flow period, the system automatically reverts to the clean-
out cycle (i.e.,
optimum parameters from continuous clean-out testing). The production testing
continues until the average daily gas rate is less than the optimum rate. The
system will
revert to operating at these optimum conditions once the testing is complete.
From a practical standpoint, there are three conditions to be avoided in
association
with sub-critical flow, as follows:
1. Accumulated liquid should not be greater than surface facilities can
handle, in
order to prevent the compressor from being flooded.
2. Liquids should not be allowed to accumulate to the point that production is
restricted.
3. Liquids must not be allowed to accumulate to the point that the compressor
does
not have capacity to remove them, or to the point that the hydrostatic
pressure of
the liquids exceeds the reservoir pressure.
Production testing will be terminated prematurely by the PLC (i.e., before the
end
of the specified test duration has been reached) if any of the above-mentioned
conditions
should develop. Otherwise, testing will continue until the end of the
specified test period
unless the well is producing less gas, in which case testing will be aborted
and the well
will revert to the previous operational parameters.
If it is desired to assess the viability of intermittent clean-out for a given
wellbore
on the basis of optimizing cash flow, definition of the production test grid
may entail the
use of inputs values for additional parameters, such as:
= Water disposal cost (per cubic meter);
= Gas price (per cubic meter); and
= Power cost (per kilowatt-hour).
-47-
CA 02714318 2010-09-08
The PLC totalizes gas and water production as well as power consumption during
each test period. These accumulations, combined with the costs and test
duration,
provide a daily cash flow indicator. Back pressure is kept constant for all
tests to ensure
that a true indication of wellbore productivity is provided. At the completion
of each
sub-critical flow period, the system automatically reverts to the clean-out
cycle (i.e.,
optimum parameters from continuous clean-out testing). Production testing
continues
until the average daily cash flow rate is less than the daily cash flow
indicator. The
system will revert to operating at these optimum conditions once the testing
is complete.
Production testing will be terminated prematurely by the PLC (i.e., before the
end
of specified test duration has been reached) if any of the above-mentioned
conditions
should develop. Otherwise testing will continue until the end of the specified
test period
unless the well is producing a reduced cash flow, in which case testing will
be aborted
and the well will revert to the previous operational parameters.
Gas lift application for an oil producer
FIGS. 12 and 13 illustrate how principles and concepts taught herein may also
be
applied to enable a simple compressor to provide onsite gas lift for a
producing oil well.
Produced gas is separated at the surface, and a portion of the produced gas re-
injected
into the wellbore through a coiled tubing string. This gas lift gas lightens
the fluid
column in the wellbore and facilitates production of liquids up the wellbore
by pressure-
induced flow (as opposed to velocity-induced flow). The well casing in this
scenario will
typically be closed, resulting in an accumulation of high-pressure gas within
the casing.
Anytime additional gas lift gas is required for the system, this high-pressure
gas stored in
the casing may be utilized for that purpose.
In the embodiment illustrated in FIG. 12, process block 300 conceptually
represents a flow-splitting apparatus. A first input flow Q1 is a flow of oil
emulsion from
a wellbore. A second input flow Q2 is gas flow from the wellbore casing. A
first output
flow Q3 is a flow of oil emulsion down a flow line. A second output flow Q4 is
a flow
of lift gas flow down the coiled tubing. It will be noted that all of the gas
lift gas passes
through the compressor.
-48-
CA 02714318 2010-09-08
A first control valve CV-1 (which may be in accordance with a control valve
embodiment disclosed elsewhere in this patent document) is controlled by a
first liquid
level sensor LLS-1 (alternatively referred to as a fluid level indicator). If
first liquid
level sensor LLS-1 is down, first control valve CV-1 will be closed. If first
liquid level
sensor LLS-1 is up, first control valve CV-1 will open. First control valve CV-
1
attempts to maintain the liquid level below first liquid level sensor LLS-1.
A second control valve CV-2 is controlled based on second liquid level sensor
LLS-2. If second liquid level sensor LLS-2 is down, second control valve CV-2
will be
closed. If second liquid level sensor LLS-2 is up, second control valve CV-2
will open.
Second control valve CV-2 attempts to maintain a liquid level above second
liquid level
sensor LLS-2.
The combined action of these two control valves attempts to maintain a liquid
level in the process vessel between the two level switches. Based on the
liquid level
sensed by a given liquid level sensor, the associated control valve will be
further opened
or closed; i.e., the control valve will pulse open in response to an increase
in the liquid
level, and closed (or less open) in response to a decrease in the liquid
level. Compressor
throughput is controlled by a variable-frequency drive (VFD).
It will be readily appreciated by those skilled in the art that various
modifications
of the present invention may be devised without departing from the scope and
teaching of
the present invention, including modifications which may use equivalent
structures or
materials hereafter conceived or developed. It is to be especially understood
that the
invention is not intended to be limited to any described or illustrated
embodiment, and
that the substitution of a variant of a claimed element or feature, without
any substantial
resultant change in the working of the invention, will not constitute a
departure from the
scope of the invention. It is to also be appreciated that the different
teachings of the
embodiments described and discussed herein may be employed separately or in
any
suitable combination to produce desired results.
In this patent document, any form of the word "comprise" is to be understood
in
its non-limiting sense to mean that any item following such word is included,
but items
-49-
CA 02714318 2010-09-08
not specifically mentioned are not excluded. A reference to an element by the
indefinite
article "a" does not exclude the possibility that more than one of the element
is present,
unless the context clearly requires that there be one and only one such
element. Any use
of any form of the terms "connect", "engage", "couple", "attach", or any other
term
describing an interaction between elements is not meant to limit the
interaction to direct
interaction between the subject elements, and may also include indirect
interaction
between the elements such as through secondary or intermediary structure.
-50-
