Note : Les descriptions sont présentées dans la langue officielle dans laquelle elles ont été soumises.
- 1 -
SYSTEM, METHOD, AND MEDIUM FOR HISTORICAL OPTIMIZATION OF OIL-
WATER SEPARATION PROCESS
FIELD
[0001] The present disclosure relates to optimizing the processing of
hydrocarbon materials, and in particular to systems, methods, and processor-
readable media for optimizing the use of chemical additives in an oil-water-
gas
separation process based on historical data.
BACKGROUND
[0002] Steam assisted gravity drainage (SAGD) is a technique for
extracting
bituminous material, such as bituminous sands (also referred to as oil sands),
from
subterranean reservoirs. Steam and/or solvent is injected into the reservoir
using
and injector well to reduce the viscosity of the bituminous material, allowing
it to be
extracted to the surface in fluid form using a producer well. In SAGD
operations,
the produced fluid is a mix of oil (in the form of bitumen), water, produced
gas, and
other materials. Separating the oil from the water occurs in stages: first, at
an inlet
separation stage, the produced fluid emulsion is separated into an oil portion
and a
water portion. Within the inlet separation stage, the oil portion first has a
majority
(e.g., 90%) of its water removed at a free water knockout (FWKO) vessel, then
the
output of the FWKO vessel has most or all of its remaining water (e.g., the
remaining 10 /0 or close to it) removed at a treater vessel, using one or more
treatments such as viscosity reduction, chemicals, etc. The oil portion from
the
treater vessel is then either sent to market, typically via pipeline or other
transportation means, or relayed for further processing.
[0003] Chemicals can be used at the inlet separation stage (such as
at the
free water knockout and treater vessels) to maximize the separation of oil and
water - typical industry standards require that the oil portion should be less
than
0.5% water before transportation by pipeline. The chemicals used can include
an
Date Recue/Date Received 2022-12-23
- 2 -
emulsion breaker (EB) to coalesce water in oil (WI0), and a reverse emulsion
breaker (REB) and coagulant to separate oil in water (OIW). These chemicals
assist
gravity separation to reduce the residence time required to affect the
separation,
i.e., the duration of time the material must remain within a given stage to
allow the
operations of that stage to achieve the desired effect.
[0004] After the inlet separation stage, the water portion undergoes
further
de-oiling and softening (also called water treatment) at a second stage.
Coagulant
can also be added at the second stage (i.e. during de-oiling and/or water
treatment) to remove residual oil from the water portion. In a typical de-
oiling
process, the water goes through a serpentine flow path through a de-oiling
tank,
such as a skim tank or a nnicrobubble flotation (MBF) tank, and may also
undergo
further de-oiling by an induced static flotation (ISF) vessel and/or an oil
removal
filter (ORF). The water portion from the de-oiling process then goes through
various
other water treatment stages to remove hardness and silica to avoid fouling
when
boiled.
[0005] After the second stage, the recovered de-oiled and softened
water is
used for steam generation. The steam is injected back into the oil reservoir
(e.g.
via an injector well), delivering heat to mobilize bitumen (i.e., to reduce
the
viscosity of the bituminous material). The water used to generate the steam
typically returns to the surface once again as a portion of the produced
fluid, and
the cycle continues.
[0006] The chemicals used at the inlet separation stage and the
second stage
are expensive - in some cases, they can account for greater than 50% of
operating
expenses for a SAGD operation. Thus, there is an incentive to optimize the
amount
of chemicals used to avoid overuse while maintaining the required
specification and
ensuring the water is ready for steam generation. Separation at both stages is
also
affected by other factors, such as residence time and diluent addition.
However,
cutting the flow rate of the oil and/or water portions (i.e. increasing their
residence
time at the various stages) can have significant costs due to reduced
production per
unit of time.
Date Recue/Date Received 2022-12-23
- 3 -
[0007] Conventionally, engineers and operators optimize the
separation
process manually. Engineers on site can monitor trends with regard to the oil
and
water content of various process streams, the chemical dosages, and the
diluent
settings. If the required specifications (e.g., low basic sediment and water
(BS&W)
concentration such as <0.5%, low OIW, low mineral content to avoid fouling)
are
satisfied, then no changes are made. However, if there is an upset (i.e. the
oil
and/or water portions fail to meet the specifications), an engineer makes a
recommendation, typically based on intuition and experience, to change the
operating parameters of the separation process. For example, an engineer may
advise that more water be removed at the inlet separation stage if the oil is
too wet
(i.e. too much water in the oil phase), even though this might make the water
portion too dirty. However, relying on the intuitions and experience of
engineers
and operators to optimize individual operating parameters of the separation
process
can result in sub-optimal and/or inconsistent performance of the process as a
whole, leading to instability and/or inefficiency.
[0008] Accordingly, it would be useful to provide techniques for
optimizing
and standardizing the operation of an oil water separation process that
overcome
one or more of the limitations identified above.
SUMMARY
[0009] The present disclosure describes systems, methods, and
processor-
readable media for optimizing the use of chemical additives in an oil-water
separation process based on historical data.
[0010] In some embodiments, a process optimization software system is
used
to optimize the operation of fluid treatment processes. The process
optimization
software system includes a model that uses historical operating data to
identify
operating conditions that are a close match to the present operating
conditions. If
the identified operating conditions are known to have given rise to a
historical
Date Recue/Date Received 2022-12-23
- 4 -
upset, the process optimization software system recommends interventions that
solved the historical upset.
[0011] In some embodiments, the model uses three categories of
operating
parameters as input variables: specified variables, control variables, and
observed
variables. The specified variables are parameters with values or ranges of
values
specified by a requirement or constraint of the process, such as a water in
oil
content range (e.g., a BS&W concentration range) for the produced oil portion
in
order to satisfy the specification for transport to market (e.g., BS&W between
0.4%
and 0.6%). The control variables are parameters that are used to identify
similar
historical states, such as flow rates of the inlet emulsion into the plant or
the water
output to the steam generation stage. The third category, observed variables,
do
not have to fall within a specified range (unlike the specified variables) and
are not
used to identify similar historical states (unlike the control variables).
Instead, the
observed parameters are simply used to provide additional information about
current and historical states to the operator of the oil water separation
system.
[0012] In some embodiments, after a favourable historical state has
been
identified and selected, one or more modifiable operating parameters of the
oil
water separation system can be modified to transition from the current state
closer
to the selected historical state. The modifiable parameters include operating
parameters that can be controlled or modified via action taken by the operator
of
the oil water separation system controlling the oil water separation process.
In
some examples, the modifiable operating parameters can include one or more of
the specified variables, control variables, and/or observed variables. For
example,
one modifiable operating parameter can denote an amount (e.g., a volume or
volumetric flow rate) of emulsion breaker to add upstream from an inlet
separation
vessel. Another example of a modifiable operating parameter is a flow rate of
a
water portion produced by an inlet separation vessel (e.g., FWKO vessel or
treater
vessel), to help maintain an oil-water interface to maintain OIW and BS&W
specifications. Because many of the variables affect each other, often in
complex
ways, it can be difficult to predict how modifying an operating parameter will
affect
Date Recue/Date Received 2022-12-23
- 5 -
other variables, unless a similar historical state can be identified in which
the new
value of the modified operating parameter coexists with other variable values.
[0013] In some embodiments, when identifying similar historical
states to the
current state, the model seeks to find a historical state that has roughly
similar
values for most control variables, and a desirable value for one or more
specified
variables. For a given point in time, each variable has a range or tolerance:
some
can have tighter tolerance than others. The model attempts to adjust one or
more
of the specified variables while remaining within the tolerance of all control
variables. Thus, in some embodiments, tolerance selection is important: the
tolerances of the control variables used by the model can be configured to be
narrow to limit how much the state is modified, but broad enough to give
useful
similar historical states.
[0014] In various embodiments, the similarity of the current state to
a
historical state is calculated using various techniques: for example, cosine
similarity
(i.e. the angle between two points), Euclidian similarity, actual distance,
Manhattan
distance, etc. Each of these techniques can exhibit various advantages and
disadvantages. In some embodiments, the similarity measure is assessed with
respect to seven dimensions or more.
[0015] In some embodiments, the model is deployed in association with
an oil
water separation system to identify best-fit matches of historical states for
the state
of the oil water separation system on the current day. In some embodiments, a
user interface can be used to present a best-fit desirable state to a human
user,
such as an engineer, along with operating parameter value adjustments to be
made
for the oil water separation system to mimic the best-fit desirable state. The
goal is
for the oil water separation system to modify the current day's operations
(i.e. to
adjust the modifiable operating parameters of the oil water separation
process) to
match the output(s) of the selected historical state, such as the amounts and
characteristics of the intermediate and final products of the oil water
separation
process. Thus, the tolerances can dictate how much the current state may need
to
Date Recue/Date Received 2022-12-23
- 6 -
be modified to match the favourable historical state in order to achieve the
output(s) of that historical state.
[0016] In some embodiments, the model is incrementally updated as
more
historical data is generated. In some embodiments, historical data older than
a
predetermined age (such as data from before a major change in the underlying
configuration of the oil water separation process) is not used by the model,
to avoid
optimizing operations based on historical states that would generate
substantially
different outputs when applied to the current system. In some embodiments,
historical data corresponding to non-typical operating circumstances, such as
emergency plant shutdown conditions, plant start-up conditions, new chemical
trials, and other atypical operating conditions can be disregarded by the
model.
[0017] Some embodiments can apply the techniques described herein to
optimize the operating parameters of a separation process other than an oil-
water
separation process. A separation system, such as an oil-water separation
system,
can be used to implement a separation process including one or more stages for
separating water from other materials. By using the techniques described
herein,
operating parameters representing a current state of the separation process
can be
modified to bring the current state of the separation process closer to a
historical
state of the separation process, thereby improving performance and/or avoiding
upset as described above.
[0018] Some example embodiments described herein can exhibit one or
more
advantages over conventional techniques, and can thereby solve one or more
technical problems. Some oil water separation systems have many years of
usable
historical operating data available to inform the operation of the model. This
historical data can be used by the model to achieve the goals of the operators
of
the oil water separation system: e.g., to adhere to the product
specification(s), to
increase product throughput, to reduce costs, to reduce waste (e.g. wasted
chemical additives, wasted energy used in heat treatment), to reduce
environmental impact, etc. By deploying a model to directly inform decision-
making
using a large body of detailed, objective historical data, the oil water
separation
Date Recue/Date Received 2022-12-23
- 7 -
process can be optimized without relying on the subjective, inconsistent,
approximate intuitions and judgment of individual engineers, which can be
imperfect when applied to a system having a large number of variables. Thus,
the
use of the model can promote consistent operating strategy based on historical
successes, i.e., standardizing and prioritizing which operating parameters to
modify. The model and process optimization software system may thereby enable
proactive daily optimization, minimize trial and error when making changes,
and
allow the identification of historical states across many more variables than
can be
comprehended simultaneously by a human engineer. Furthermore, the model can
improve its recommendations over time by continuously collecting data from
different operating states to add to its body of historical data.
[0019] In the present disclosure, the term "product" refers to either
an
intermediate product or a final product of an oil-water separation system.
[0020] As used herein, the term "threshold" refers to a limit on a
value. The
threshold may be a lower limit, an upper limit, an absolute limit of absolute
magnitude, or a relative limit with respect to the current value of a system
variable,
or any other limit. Statements that a value is "within" a threshold refer to
the value
being within a region or range bounded by the threshold.
[0021] In the present disclosure, the terms "parameter" and
"variable" are
used interchangeably to refer to a measurable aspect or characteristic of a
system.
Parameters have values, which can be scalar, vector, or other types of values,
and
can change over time.
[0022] As used herein, statements that a second item (e.g., a signal,
value,
scalar, vector, matrix, calculation, or bit sequence) is "based on" a first
item can
mean that characteristics of the second item are affected or determined at
least in
part by characteristics of the first item. The first item can be considered an
input to
an operation or calculation, or a series of operations or calculations, that
produces
the second item as an output that is not independent from the first item.
Date Recue/Date Received 2022-12-23
- 8 -
[0023] In some aspects, the present disclosure describes a method
that
includes obtaining current state data comprising a plurality of operating
parameter
values for a separation process, the separation process comprising one or more
stages for separating water from other materials, the plurality of operating
parameter values jointly representing a current state of the separation
process. The
method further includes obtaining historical data comprising a plurality of
historical
data samples, each historical data sample representing a respective historical
state
of the separation process comprising a plurality of historical operating
parameter
values. The current state data and the historical data are processed to
identify one
or more historical states that satisfy a similarity condition with respect to
the
current state and that also satisfy a desirability condition. A selected
historical state
from the one or more historical states is applied to the separation process by
modifying current values of one or more of the operating parameters of the
separation process based on respective historical operating parameter values
indicated by the selected historical state.
[0024] In some example aspects, the similarity condition requires
each
operating parameter value of the current state to fall within a respective
tolerance
threshold of a corresponding historical operating parameter value of the
historical
state.
[0025] In one or more of the preceding aspects, the similarity condition
requires that a distance value, measured between the plurality of operating
parameter values of the current state and the respective plurality of
historical
operating parameter values of the historical state, is within a distance
threshold.
[0026] In one or more of the preceding aspects, the distance value is
calculated based on at least one of: cosine similarity, Euclidian distance,
actual
distance, and Manhattan distance.
[0027] In one or more of the preceding aspects, the desirability
condition
requires that a water concentration parameter of the operating parameters,
indicating a water concentration of a water portion produced by a stage of the
one
or more stages, is within a water concentration range.
Date Recue/Date Received 2022-12-23
- 9 -
[0028] In one or more of the preceding aspects, the desirability
condition
requires that a non-water substance concentration parameter of the operating
parameters, indicating a concentration of a specific non-water substance of a
non-
water portion produced by a stage of the one or more stages, is within a non-
water
substance concentration range.
[0029] In one or more of the preceding aspects, the desirability
condition
requires that a chemical additive amount parameter of the operating
parameters,
indicating an amount of a chemical additive added to a stage of the one or
more
stages, is within a chemical additive range.
[0030] In one or more of the preceding aspects, the separation process
comprises an oil water separation process comprising one or more stages for
separating water from oil.
[0031] In one or more of the preceding aspects, the desirability
condition
requires that a water concentration parameter of the operating parameters,
.. indicating a water concentration of a water portion produced by a stage of
the one
or more stages, is within a water concentration range.
[0032] In one or more of the preceding aspects, a stage comprises an
inlet
separation stage.
[0033] In one or more of the preceding aspects, a stage comprises a
de-oiling
.. stage.
[0034] In one or more of the preceding aspects, the desirability
condition
requires that an oil concentration parameter of the operating parameters,
indicating
a concentration of oil of an oil portion produced by a stage of the one or
more
stages, is within an oil concentration range.
[0035] In one or more of the preceding aspects, a stage comprises an inlet
separation stage.
[0036] In one or more of the preceding aspects, the desirability
condition
requires that a chemical additive amount parameter of the operating
parameters,
Date Recue/Date Received 2022-12-23
- 10 -
indicating an amount of a chemical additive added to a stage of the one or
more
stages, is within a chemical additive range. In one or more of the preceding
aspects, the third stage comprises an inlet separation stage; and the chemical
additive comprises one of the following: an emulsion breaker; a reverse
emulsion
breaker; and a coagulant.
[0037] In one or more of the preceding aspects, the chemical additive
comprises a coagulant.
[0038] In one or more of the preceding aspects, modifying the current
values
of the one or more of the operating parameters of the separation process
comprises
modifying the current value of at least one of the following operating
parameters:
a flow rate of a product of a stage of the plurality of stages; a temperature
of a
stage of the plurality of stages; an amount of a diluent to add to a stage of
the
plurality of stages; an amount of a diluent to add to a product of a stage of
the
plurality of stages; an amount of an emulsion breaker to add to a stage of the
plurality of stages; an amount of a reverse emulsion breaker to add to a stage
of
the plurality of stages; and an amount of a coagulant to add to a stage of the
plurality of stages.In some aspects, the present disclosure describes a
system. The
system comprises a processor device, and a memory storing instructions that,
when
executed by the processor device, cause the system to perform one or more of
the
methods described above. ,
[0039] In some aspects, the present disclosure describes a non-
transitory
computer-readable medium storing instructions thereon to be executed by a
processor device, the instructions, when executed, causing the processor
device to
perform one or more of the methods described above.
BRIEF DESCRIPTION OF THE DRAWINGS
[0040] Reference will now be made, by way of example, to the
accompanying
drawings which show example implementations of the present application, and in
which:
Date Recue/Date Received 2022-12-23
- 11 -
[0041] FIG. 1 is a block diagram showing the stages of a SAGD system
suitable for implementation of examples described herein.
[0042] FIG. 2 is a block diagram of an example computing system
suitable for
implementation of examples described herein.
[0043] FIG. 3 is a block diagram of an example process optimization
software
system, in accordance with example implementations described herein.
[0044] FIG. 4A is a graph of similar historical states to a current
state with
respect to two specified variables, using a Euclidean similarity measure, in
accordance with example implementations described herein.
[0045] FIG. 4B is a graph of similar historical states to a current state
with
respect to two specified variables, using a cosine similarity measure, in
accordance
with example implementations described herein.
[0046] FIG. 5 is a flowchart showing operations of a method for
optimizing the
operation of an oil water separation system using historical data, in
accordance with
example implementations described herein.
[0047] Similar reference numerals can have been used in different
figures to
denote similar components.
DESCRIPTION OF EXAMPLE IMPLEMENTATIONS
[0048] The present disclosure describes systems, methods, and processor-
readable media for optimizing the use of chemical additives and separation
vessels
interface control in an oil-water separation process based on historical data.
Systems and methods will be described with reference to a process optimization
software system used to control or assist the operation of an oil water
separation
process of a SAGD system. However, it will be appreciated that the techniques
described herein can also be applied to other separation processes that
separate
water from other substances, and/or to the optimization of other operating
Date Recue/Date Received 2022-12-23
- 12 -
parameters of such processes, and/or can be implemented through other
computational means.
[0049] FIG. 1 is a block diagram of an example SAGD system 100,
including
an extraction system 110 and an oil water separation system 120. The
extraction
system 110 includes one or more injector wells 114 and one or more producer
wells
112. The injector well(s) 114 are configured to inject steam 168 received from
the
steam generation stage 142, potentially mixed with other injected fluids 14
(such
as solvents) to form an injected fluid 12, into a subterranean reservoir 10 of
bituminous material such as oil sands. In operation, the injected fluid 12
mobilizes
the bituminous material and causes it to flow toward the bottom producer
well(s)
112, which are used to transport the bituminous material, now mixed with water
and/or other fluids from the injected fluid 12 as a fluid emulsion 152, to the
surface. The fluid emulsion 152 is provided to the oil water separation system
120
to produce oil from the bitumen content of the fluid emulsion 152 and to
recycle the
water content of the fluid emulsion 152 for use as steam 168.
[0050] The stages of the oil water separation system 120 will now be
described briefly. It will be appreciated that the stages shown in FIG. 1 are
functional units that can include many sub-components such as pumps, heaters,
and other equipment not fully described herein.
[0051] An inlet separation stage 122 includes one or more free water
knockout (FWKO) vessels 124 and one or more treater vessels 126. The FWKO
vessel 124 receives the fluid emulsion 152, potentially diluted with a diluent
178,
and removes most of the water from the (diluted or non-diluted) fluid emulsion
152
as a FWKO water portion 156. The water can be removed by gravity separation;
the
duration of time required to effect the gravity separation can vary based on
various
operating parameters, but can sometimes be shortened by addition of FWKO
chemical additives 170 upstream of the FWKO vessel(s) 124, which can include
an
emulsion breaker (EB) to coalesce the water in oil (WIO) more effectively,
and/or a
reverse emulsion breaker (REB) and/or a coagulant to separate oil in water
(OIW)
more effectively. Thus, use of greater amounts of some chemical additives 170
can
Date Recue/Date Received 2022-12-23
- 13 -
potentially reduce the FWKO residence time necessary to affect a given amount
of
emulsion in the FWKO vessel 124.
[0052] The FWKO oil portion 154 is received by the one or more
treater
vessels 126. The additional water removed from the FWKO oil portion 154 by the
treater vessel 126 constitutes the treater water portion 160; the treater oil
portion
158 is a final product output by the oil water separation system 120 as oil
ready for
storage, transportation, or further processing.
[0053] A de-oiling vessel 134 and water treatment stage 136 operate
to
further process the FWKO water portion 156 and the treater water portion 160
(which may be combined into a single stream in some embodiments) to prepare
the
water for conversion to steam 168 for use in further SAGD extraction
operations. A
de-oiling vessel 134, such as a skim tank, receives the FWKO water portion 156
and the treater water portion 160, de-oiling the water portions 156, 160 by
mechanically removing floating oil. De-oiling tank chemical additives 174,
such as
coagulants, can be used to improve the effectiveness of the de-oiling
operation of
the skim tank 134. In some embodiments, the skim tank oil portion 162 can be
provided as a final product for storage or transport along with the treater
oil portion
158. The de-oiling tank water portion 164 is provided to a water treatment
stage
136 for softening (i.e. removal of mineral content) and/or removal of silica.
In some
.. embodiments, the de-oiling tank water portion 164 undergoes further de-
oiling,
such as by an induced static flotation (ISF) vessel and/or an oil removal
filter
(ORF), before proceeding to the water treatment stage 136. In some
embodiments,
the softening process of the water treatment stage 136 can be assisted by
addition
of water treatment chemical additives 176. The treated water 166 produced by
the
water treatment stage 136 is provided to a steam generation stage 142, which
uses
a boiler 144 to generate steam 168 from the treated water 166. The steam 168
is
provided to the extraction system 110.
[0054] The process conditions of each unit operation of the oil water
separation system 120, i.e. each stage, can be defined by a set of parameters
referred to herein as operating parameters. Some operating parameters can be
Date Recue/Date Received 2022-12-23
- 14 -
controlled or modified by an operator of the oil water separation system 120,
and
can be referred to herein as modifiable operating parameters. Adjustment of a
modifiable operating parameter can include adjustment of actual physical
conditions
of a process within the oil water separation system 120, through various
physical
adjustment means, such as actuating valves, modulating the amount of energy
supplied to a heating element, instructing personnel to increase or decrease
an
amount of chemical additive manually added to a stage, etc. In various
embodiments, the modifiable operating parameters can include flow rates of
various products 154, 156, 158, 160, 162, 164, flow rates of various diluents
178,
flow rates of various chemical additives 170, 174, 176, temperatures of
various
stages, etc.
[0055] Other operating parameters can be measured (e.g., by sensors),
but
not directly controlled.
[0056] FIG. 2 is a block diagram of an example computing system 240
including computing hardware suitable for optimizing the operation of an oil
water
separation system 120 according to example embodiments described herein. In
some implementations, computing system 240 can be an electronic computing
device, such as a networked server. In other implementations, the computing
system 240 can be a distributed computing system including multiple devices
(such
.. as a cloud computing platform) or a virtual machine running on one or more
devices in mutual communication over a network. Other examples suitable for
implementing implementations described in the present disclosure can be used,
which can include components different from those discussed below. Although
FIG.
2 shows a single instance of each component, there can be multiple instances
of
each component in the computing system 240.
[0057] The computing system 240 can include one or more processor
devices
(collectively referred to as processor device 242 or processor 242). The
processor
device 242 can include one or more processor devices such as a processor, a
microprocessor, a digital signal processor, an application-specific integrated
circuit
Date Recue/Date Received 2022-12-23
- 15 -
(ASIC), a field-programmable gate array (FPGA), a dedicated logic circuitry, a
dedicated artificial intelligence processor unit, or combinations thereof.
[0058] The computing system 240 can include one or more network
interfaces
(collectively referred to as network interface 246) for wired or wireless
communication over a network. The network interface 246 can include wired
links
(e.g., Ethernet cable) and/or wireless links (e.g., one or more antennas). The
computing system 240 can communicate with one or more user devices 247 (such
as user workstation computers) via the network interface 246. The computing
system 240 can also communicate with various sensors 248 (e.g. flow rate
sensors,
temperature sensors, pH sensors, visual sensors) or other data sources (e.g.,
lab
test results) to obtain data used in operating and optimizing the oil water
separation system 120. The computing system 240 can also communicate with
various process controllers 249 via the network interface 246 to control the
modifiable operating parameters of the various components of the oil water
separation system 120. In some examples, the user devices 247 and/or the
components of the oil water separation system 120 can communicate with the
computing system 240 through other means, such as an input/output interface of
the computing system 240 (not shown) or through an intermediate device in
communication with the computing system 240.
[0059] The computing system 240 can include one or more non-transitory
memories (referred to collectively as a memory 244), which can include a
volatile
or non-volatile memory (e.g., a flash memory, a random access memory (RAM),
and/or a read-only memory (ROM)). The memory 244 can also include one or more
mass storage units, such as a solid state drive, a hard disk drive, a magnetic
disk
drive and/or an optical disk drive.
[0060] The memory 244 can store instructions for execution by the
processor
device 242 to carry out examples described in the present disclosure. The
instructions can include instructions for implementing and operating the
process
optimization software system 300 described below with reference to FIG. 3. In
some embodiments, the process optimization software system 300 includes
Date Recue/Date Received 2022-12-23
- 16 -
subsystems or functional modules such as a model 302 and a user interface (UI)
module 304, both described below with reference to FIG. 3. The memory 244 can
include other software instructions, such as for implementing an operating
system
and other applications/functions. In some examples, the computing system 240
can
additionally or alternatively execute instructions from an external memory
(e.g., an
external drive in wired or wireless communication with the computing system
240)
or can be provided executable instructions by a transitory or non-transitory
computer-readable medium. Examples of non-transitory computer readable media
include a RAM, a ROM, an erasable programmable ROM (EPROM), an electrically
erasable programmable ROM (EEPROM), a flash memory, a CD-ROM, or other
portable memory storage.
[0061] The memory 244 can store data 270 used by the process
optimization
software system 300. Current state data 310 can be received from the sensors
248,
process controllers 249, and/or other data sources and stored in the memory.
The
.. current state data 310 can include values for one or more specified
variables 312,
control variables 314, and/or observed variables 316, as described above.
Historical
data 320 can also be stored, consisting of a plurality of historical state
data 321
records, each of which includes values for one or more specified variables
322,
control variables 324, and/or observed variables 326. The current state data
310,
and each historical state data 321 record, represents a single state of the
oil water
separation process implemented by the oil water separation system 120.
[0062] The computing system 240 can also include a bus 250 providing
communication among components of the computing system 240, including those
components discussed above. The bus 250 can be any suitable bus architecture
including, for example, a memory bus, a peripheral bus or a video bus, or the
bus
250 can be another communication link such as a network interface 246.
[0063] FIG. 3 illustrates an example process optimization software
system
300. The process optimization software system 300 is executed by a computing
system 240 to perform the methods and operations described herein. The process
.. optimization software system 300 includes functional modules or subsystems
302,
Date Recue/Date Received 2022-12-23
- 17 -
304, as described below. It will be appreciated that some implementations can
omit
one or more of the described subsystems and/or can combine the functions of
two
or more of the described subsystems into a single component. In some
implementations, different functions of the process optimization software
system
300 can be performed on different devices other than the computing system 240.
For example, computationally intensive functions such as training machine
learning
models and executing trained machine learning models can be performed on a
cloud computing platform in communication with a local computing system 240.
[0064] In some implementations, the process optimization software
system
300 operates to identify desirable historical states of the oil water
separation
system 120 that are similar to the current state, and to modify or recommend
modifications to the operating parameters of the current state to more closely
approximate a selected historical state in order to stabilize and/or optimize
the
operation of the oil water separation system 120. The operation of the process
optimization software system 300 will be described with reference to FIG. 3 as
well
as the flowchart of FIG. 5 and the similarity measures of FIG.s 4A-4B.
[0065] FIG. 5 is flowchart showing operations of a method 500 for
optimizing
the operation of an oil-water separation process based on historical data. The
method 500 will be described in the context of the example oil water
separation
system 120, using a computing system 240 executing the process optimization
software system 300. However, it will be appreciated that the operations or
steps of
method 500 are not limited to this example and can be implemented using other
separation processes or systems, using other software systems executed by
other
computing systems, etc.
[0066] At 502, current state data 310 is obtained. As described above, the
current state data 310 includes one or more specified variables 312, control
variables 314, and/or observed variables 316 that are determined from data
received from the sensors 248, the process controllers 249, and/or other data
sources (not shown) such as laboratory testing results, predetermined fixed
values
of various equipment, etc.
Date Recue/Date Received 2022-12-23
- 18 -
[0067] The specified variables 312 have known specifications used as
criteria
for identifying valid historical states, i.e., historical states that satisfy
a desirability
condition. In some examples, the specified variables 312 include a basic
sediment &
water (BSW) variable indicating the concentration of water and sediment in a
final
oil product such as 158 or 162 (e.g., having an example specification of
between
0.4% and 0.6%). In some examples, the specified variables 312 include a dilbit
density variable indicating the density of dilbit (i.e. diluted bitumen); an
example
specification for dilbit density can be between 915 and 920 kg/m3. In some
examples, the dilbit density can be that at the gas boot (i.e. at the output
of the oil
water separation system 120, where final oil products 158 undergo a final de-
gassing step not shown in FIG. 1). In some examples, the specified variables
312
include an emulsion breaker (EB) dosage variable indicating a concentration of
EB
of the FWKO chemical additives 170 injected upstream of the FWKO vessel 124.
In
some embodiments, the historical data 320 is filtered to remove historical
state
data 321 records not meeting the specifications. In some embodiments, the
specifications 350 are provided by a subject matter expert or by product
requirements. In various examples, the specifications 350 may be coded into
the
process optimization software system 300 and/or received from external data
sources or users via the user devices 247 and/or the network interface 246 for
storage in the memory 244.
[0068] The control variables 314 do not have a specification 350, but
are used
to identify historical states (based on their respective historical state data
321
records) that are similar to the current state (based on the current state
data 310),
i.e., historical states that satisfy a similarity condition. In some examples,
the
control variables 314 include produced water measurement variables at the FWKO
vessel 124 and the treater vessel(s) 126 (e.g., flow rates of the FWKO water
portion 156 and/or the treater vessel water portion 160). In some examples,
the
control variables 314 include an inlet emulsion flow rate variable (e.g., the
flow rate
of fluid emulsion 152).
Date Recue/Date Received 2022-12-23
- 19 -
[0069] The observed variables 316 are not compared to a specification
350 to
satisfy a desirability condition, nor are they used to identify similar
historical states
that satisfy a similarity condition. Instead, after the model 302 identifies
historical
states meeting the specifications (i.e., the specified variables 322 of the
historical
state data 321 satisfy the desirability condition) that are similar to the
current state
of the plant (i.e., the control variables 324 of the historical state data 321
satisfy
the similarity condition), the UI module 304 also presents the values of other
variables associated with the identified historical states. The other
variables whose
values are displayed to a user by the UI model 304 are the observed variables
316,
326. In some examples, the observed variables 316 include a water cut setpoint
(i.e., the ratio of the water which is produced from the producer well(s) 112
compared to the volume of the total liquids produced, i.e., the water
concentration
of the fluid emulsion 152). In some examples, the observed variables 316
include
an emulsion breaker (EB) set point, i.e., the baseline amount (e.g.,
volumetric flow
rate or concentration) of EB added to the inlet separation stage 122 in steady-
state
operation.
[0070] At 504, historical data 320 is obtained. The historical data
320 includes
multiple historical state data 321 records, each of which includes one or more
specified variables 322, control variables 324, and/or observed variables 326
collected during historical operation of the oil water separation system 120.
In
some examples, the historical data 320 can be supplemented with data collected
from other oil water separation systems that are identical or nearly-identical
to the
oil water separation system 120.
[0071] At 506, the model 302 processes the current state data 310 and
the
historical data 320 to identify one or more historical states that satisfy a
similarity
condition with respect to the current state and that also satisfy a
desirability
condition.
[0072] In some examples, the similarity condition is satisfied by
historical
states in which the value of each control variable and/or specified variable
falls
within a respective tolerance threshold of its value in the current state,
determined
Date Recue/Date Received 2022-12-23
- 20 -
according to a similarity measure. In some embodiments, the similarity
condition is
satisfied by historical states in which a distance value, measured between the
values of the control variables and/or specified variables of the current
state and
the respective values of the control variables and/or specified variables of
the
historical state, is within a distance threshold. In some embodiments, the
distance
value can be calculated using various similarity measures, such as cosine
similarity,
Euclidian distance, actual distance, and Manhattan distance. Two example
similarity
measures, cosine similarity and Euclidian distance, will now be described with
reference to FIG.s 4A-4B.
[0073] FIG. 4A shows the determination of a Euclidean distance measure in a
simplified example using only two variables, a first specified variable sv1
and a
second specified variable 5v2, to compute the similarity measure between a
current
state and a set of historical states. FIG. 4A shows a graph 400 having the
value of
sv1 as the X axis 402 and the value of 5v2 as the Y axis 404. The value of
(sv1,
5v2) of the current state data 310 is indicated by data point 410. The model
302
applies a Euclidean distance measure to identify historical state data 321
records
that satisfy the similarity condition. The values of (sv1, 5v2) of each
historical state
data 321 record are graphed as points 440. By applying the Euclidean distance
measure, data points having the shortest straight-line distance from point 410
are
.. determined to satisfy the similarity condition: namely, similar points 420.
Of the
similar points 420, a desirability measure may be applied in some embodiments
to
select a similar point 420 with the most desirable value for sv1 and/or 5v2,
in this
example desirable similar point 430.
[0074] FIG. 4B shows the determination of a cosine similarity measure
in the
.. same simplified example as FIG. 4A. The value of (sv1, 5v2) of the current
state
data 310 is indicated by data point 410. The model 302 applies a cosine
similarity
measure to identify historical state data 321 records that satisfy the
similarity
condition. The values of (sv1, 5v2) of each historical state data 321 record
are
graphed as points 440. By applying the cosine similarity measure, the
direction of
point 410 from the origin (sv1=0, 5v2=0) is determined (shown as ray 460), and
Date Recue/Date Received 2022-12-23
- 21 -
data points in this direction (i.e. close to ray 460) are determined to
satisfy the
similarity condition: namely, similar points 420. Of the similar points 420, a
desirability measure may be applied in some embodiments to select a similar
point
420 with the most desirable value for sv1 and/or 5v2, in this example
desirable
similar point 430.
[0075] As will be appreciated from FIG.s 4A and 4B, the set of data
points
that are similar to an observed (current state) data point can be quite
different
depending on the measure of similarity used. The differences can be more
pronounced with a high number of input variables. For example, in some SAGD
systems, similar states can be identified (i.e. the similarity condition can
be
satisfied) by comparing the values of 8 specified variables and 17 control
variables.
[0076] Returning to step 506 of method 500, the desirability
condition can be
satisfied in various embodiments through a combination of hard constraints
and/or
desirability metrics for the specified variables 312, 322. Some embodiments
use a
desirability condition that requires each specified variable 312, 322 to
remain within
a specified range (i.e., hard constraints). Some embodiments use a
desirability
condition that requires an overall desirability measure to fall within a
range,
wherein the desirability measure is a function of the current values of the
specified
variables (i.e., soft constraints). Some embodiments use a desirability
condition
that combines hard constraints and soft constraints for various specified
variables.
[0077] The historical states whose historical state data 321
satisfies the
similarity condition and the desirability condition are added to a list of
identified
historical states. In some embodiments, a historical state is selected
algorithmically
from the list of identified historical states by the process optimization
software
.. system 300. In some embodiments, a state that is most similar to the
current
state of the plant, but with more optimal values for one or more specified
variables,
is selected.
[0078] At 508, the selected historical state is presented to a user.
The UI
module 304 presents the selected identified historical state, e.g. via the
network
interface 246 to user device 247. In some embodiments, the selected historical
Date Recue/Date Received 2022-12-23
- 22 -
state can be presented on a display as a set of variable values of the
historical state
data 321, including not only values of the specified variables 322, but also
of the
control variables 324 and/or observed variables 326 in order to display the
full
description of the selected historical state to the user. In some embodiments,
the
user is also presented with the point in time in the oil water separation
system's
120 history from which the selected historical state was drawn, so that the
user can
understand more about the relevance and recency of the selected historical
state.
[0079] At 510, user input is received (e.g., through the user device
247 via
the network interface 246) confirming the selection of the selected historical
state
(e.g., the selected historical state 340) for emulation by the oil water
separation
system 120.
[0080] In some embodiments, steps 508 and/or 510 are omitted, and
user
confirmation is not required; the process optimization software system 300
automatically applies the selected historical state to the oil water
separation system
120.
[0081] At 512, the selected historical state 340 is applied to the
oil water
separation process of the oil water separation system 120. The current values
of
one or more of the modifiable operating parameters of the oil water separation
system 120 are modified, based on the respective values of the modifiable
operating parameter values indicated by the selected historical state 340. In
some
examples, some or all of the modifiable operating parameters are modified
directly
via the process controller 249. In some examples, some or all of the
modifiable
operating parameters are modified indirectly via intervention by workers. The
new,
modified values of the modifiable operating parameters are used to update the
current state data 310 to reflect the new, modified current state.
[0082] At 514, the oil water separation system 120 is operated, in
accordance
with the newly modified current state. By making these modifications, the
current
state of the oil water separation system 120 becomes a state that is likely to
satisfy
the specification(s) 350, e.g., BSW within a certain tolerance, more optimal
chemical usage, etc.
Date Recue/Date Received 2022-12-23
- 23 -
General
[0083] Although the present disclosure describes functions performed
by
certain components and physical entities, it should be understood that, in a
distributed system, some or all of the processes can be distributed among
multiple
components and entities, and multiple instances of the processes can be
carried out
over the distributed system.
[0084] Although the present disclosure describes methods and
processes with
steps in a certain order, one or more steps of the methods and processes can
be
omitted or altered as appropriate. One or more steps can take place in an
order
other than that in which they are described, as appropriate.
[0085] Although the present disclosure is described, at least in
part, in terms
of methods, a person of ordinary skill in the art will understand that the
present
disclosure is also directed to the various components for performing at least
some
of the aspects and features of the described methods, either by way of
hardware
components, software or any combination of the two. Accordingly, the technical
solution of the present disclosure can be embodied in the form of a software
product. A suitable software product can be stored in a pre-recorded storage
device
or other similar non-volatile or non-transitory computer readable medium,
including
DVDs, CD-ROMs, USB flash disk, a removable hard disk, or other storage media,
for
example. The software product includes instructions tangibly stored thereon
that
enable a processing device (e.g., a personal computer, a server, or a network
device) to execute examples of the methods disclosed herein. In general, the
software improves the operation of the hardware in one or more ways.
[0086] The present disclosure can be embodied in other specific forms
without
departing from the subject matter of the claims. The described example
implementations are to be considered in all respects as being only
illustrative and
not restrictive. Selected features from one or more of the above-described
implementations can be combined to create alternative implementations not
explicitly described, features suitable for such combinations being understood
within the scope of this disclosure.
Date Recue/Date Received 2022-12-23
- 24 -
[0087] All values and sub-ranges within disclosed ranges are also
disclosed.
Also, although the systems, devices and processes disclosed and shown herein
can
include a specific number of elements/components, the systems, devices and
assemblies could be modified to include additional or fewer of such
elements/components. For example, although any of the elements/components
disclosed can be referenced as being singular, the implementations disclosed
herein
could be modified to include a plurality of such elements/components. The
subject
matter described herein intends to cover and embrace all suitable changes in
technology.
Date Recue/Date Received 2022-12-23
