STIMULATED WELLS

FIELD

[0001] The present application relates to the field of steam stimulated hydrocarbon production technologies, and particularly to methods and systems for steam assisted gravity drainage (SAGD) wells.

BACKGROUND

[0002] In steam assisted gravity drainage, the production of a well is dependent on geological characteristics of the well as well as the age of the well and amount of steam supplied to the well.

[0003] When developing large resources with numerous wells, the performance of individual wells are often independently monitored by different engineers.

SUMMARY

[0004] In accordance with one aspect, there is provided a system for controlling operation of a plurality of wells. The system includes a plurality of control devices for adjusting the operational inputs of the plurality of wells; a plurality of input devices for measuring well conditions and production rates; and a controller having at least one processor. The at least one processor is configured for: generating models for each of the plurality of wells based on historical well data from the plurality of input devices, the models mapping a production rate based on at least one operational input; based on one or more defined total operational constraints across all of the plurality of wells and the models, determining a distribution of operational inputs across the plurality of wells or well portions which results in an optimal total production rate; and generating signals for applying, at the plurality of control devices, the operational inputs to the wells or portions of the wells in accordance with the determined distribution.

[0005] In accordance with one aspect, there is provided a method of controlling operation of a plurality of wells. The method includes: generating, with at least one processor, models for each of the plurality of wells based on historical well data, the models mapping a production rate based on at least one operational input; based on one or more defined total operational constraints across all of the plurality of wells and the models, determining a distribution of operational inputs across the plurality of wells or well portions which results in an optimal total production rate ; and applying the operational inputs to the wells or portions of the wells in accordance with the determined distribution.

[0006] Many further features and combinations thereof concerning embodiments described herein will appear to those skilled in the art following a reading of the present disclosure.

DESCRIPTION OF THE FIGURES

[0007] In the figures,

[0008] Fig. 1A is a cross sectional view of an example geological formation and SAGD well;

[0009] Fig. 1 B is a top view of a geological area illustrating SAGD wells and infrastructure for an example project;

[0010] Fig. 2 is an example system to which aspects of the present disclosure may be applied; and

[001 1 ] Fig. 3 is a flowchart illustrating aspects of an example method of controlling operations of a plurality of wells. DETAILED DESCRIPTION

[0012] Fig. 1A shows an example of a steam assisted gravity drainage (SAGD) well 155 in a geological resource 1 10. In SAGD, production is typically effected by a pair of wells 155: an injector well 150 for injecting steam into the geological formation, and a producer well 160 for collecting the resulting bitumen. [0013] Fig. 1 B shows a top elevation view of a geological resource 1 10 having many wells (pairs) 155. The well(s) may be part of one or more SAGD projects for extracting the hydrocarbon resources in the geological formation. As illustrated by the example project in Fig. 1 B, these projects may have any number of wells 155 having any number of orientations and locations. The project(s) may include one or more facilities 120 such as well pads, plants, water sources, control systems, monitoring systems, steam generators, upgraders and any other infrastructure for extracting and/or processing input and output materials. [0014] The wells and/or infrastructure can include one or more input devices 130 for measuring, detecting or otherwise collecting data regarding the wells and processes. This data can, in some examples, include well conditions and output or production rates.

[0015] In some examples, the input devices 130 can include thermocouples or other temperature sensors, pressure sensors, and the like for measuring temperature, pressure and/or other conditions within the wells, proximate to the wells, and/or at the surface. In some examples, multiple input devices can be positioned along the length of the wells to measuring well conditions at various points in or around the length of the wells. For example, pressure and/or temperature sensors may be positioned at the toe of the well, the heel of the well, at the surface and/or elsewhere in the project infrastructure. In some examples, input sensors from reference wells, surrounding production wells, or other wells may also provide well condition information for a proximate well.

[0016] In some examples, inputs devices 130 may include flow sensors at the surface, at positions along the well and/or within any other project infrastructure to provide flow information and/or bitumen production rates. In some examples, input devices 130 can include sensors, measuring devices, and/or computational devices for determining a well's production rates of a desired hydrocarbon after processing and/or removal of water and/or other materials. In some examples, the devices may include flow meters for measuring total fluid extracted from the well.

[0017] The wells and/or infrastructure can include one or more control devices 140 for adjusting the operational inputs of the wells. In some examples, these control devices 140 can include valves, pumps, mixers, boilers, nozzles, sliding sleeves, inflow/injection control devices, drives, motors, relays and/or any other devices which may control or affect the operational inputs of the wells. In some examples, these control device(s) 140 may be configured, controlled or otherwise adjusted to change operational inputs via signals or instructions received from one or more processors in the system. For example, one or more of the control devices 140 may include controllers, processors, communication devices, electrical switches and/or other circuitry, devices or logic which can be configured, instructed or otherwise triggered to change operational inputs such as steam injection rates, temperatures, pressures, steam injection locations, pump speeds, water consumption rates, fuel consumption and any other adjustable or controllable aspect of the system. In some examples, one or more of the control devices 140 may be additionally or alternatively controlled by physical mechanisms. [0018] The number and location of the input devices 130 and control devices 140 in Figs. 1A and 1 B are illustrative examples only as any number, location and/or type of these devices 130, 140 is possible.

[0019] In some example embodiments, the input devices 130 can include sensing device cables/wires which may run the length of an entire well or portion of a well, and may provide continuous or spaced measurements along the length of the cable/wire.

[0020] In SAGD, the rate at which bitumen and hydrocarbon materials are produced from a well is dependent on various factors including, but not limited to, steam injection rates, pressure, temperature, steam chamber size/shape and properties, and physical properties of the reservoir. In some instances, too much or too little steam, pressure, temperature or other operational inputs/constraints may negatively and/or permanently affect the current and future production of a well.

[0021] During production, control device(s) 140 can be adjusted, for example, to affect a maximum predicted production rate or a desired production rate which may maximize the lifespan and total production for the well.

[0022] Individual wells may be monitored and/or managed by different individuals to produce optimal results for each well. Data driven simulations can be performed by applying initial conditions and parameters for every 3D cell and adjusting them over time based on differentials and thermal properties. These simulations can be computationally intensive, may take weeks of an expensive and powerful computing device's time before results are obtained.

[0023] However, in some situations, simulations may not provide timely enough information to react to the changing nature of steam chambers and SAGD production factors, and may not take operational constraints or trade-offs into consideration.

[0024] In some example embodiments, systems and methods described herein may provide models which can, in some instances, provide timely production rates, and may determine operational inputs which may provide optimal or improved production rates based on constraints on an individual well as well as global constraints for a large number of wells or system.

[0025] For example, in some instances, resource production at a site/project/resource/system/number of wells may have limited water supply and/or steam producing capabilities. In another example, there may be a global constraint on the rate at which the extracted hydrocarbon/water mixture can be processed, or the amount of carbon dioxide which can be generated.

[0026] In some example embodiments, systems and methods described herein may apportion limited resources/constraints for an entire system/set of wells to achieve optimal or improved production results whiles meeting these constraints.

[0027] The embodiments of the devices, systems and methods described herein may be implemented in a combination of both hardware and software. These embodiments may be implemented on programmable computers, each computer including at least one processor, a data storage system (including volatile memory or non-volatile memory or other data storage elements or a combination thereof), and at least one communication interface.

[0028] Program code may be applied to input data to perform the functions described herein and to generate output information. The output information may be applied to one or more output devices. In some embodiments, the communication interface may be a network communication interface. In embodiments in which elements may be combined, the communication interface may be a software communication interface, such as those for interprocess communication. In still other embodiments, there may be a combination of communication interfaces implemented as hardware, software, and combination thereof. In some examples, devices having at least one processor may be configured to execute software instructions stored on a computer readable tangible, non-transitory medium.

[0029] The following discussion provides many example embodiments. Although each embodiment represents a single combination of inventive elements, other examples may include all possible combinations of the disclosed elements. Thus if one embodiment comprises elements A, B, and C, and a second embodiment comprises elements B and D, other remaining combinations of A, B, C, or D, may also be used.

[0030] The technical solution of embodiments may be in the form of a software product. The software product may be stored in a non-volatile or non-transitory storage medium, which can be a compact disk read-only memory (CD-ROM), a USB flash disk, or a removable hard disk. The software product includes a number of instructions that enable a computer device (personal computer, server, or network device) to execute the methods provided by the embodiments. [0031] The embodiments described herein are implemented by physical computer hardware, including computing devices, servers, receivers, transmitters, processors, memory, displays, and networks. The embodiments described herein provide useful physical machines and particularly configured computer hardware arrangements. The embodiments described herein are directed to electronic machines and methods implemented by electronic machines adapted for processing and transforming electromagnetic signals which represent various types of information. The embodiments described herein pervasively and integrally relate to machines, and their uses; and the embodiments described herein have no meaning or practical applicability outside their use with computer hardware, machines, and various hardware components. Substituting the physical hardware particularly configured to implement various acts for non-physical hardware, using mental steps for example, may substantially affect the way the embodiments work. Such computer hardware limitations are clearly essential elements of the embodiments described herein, and they cannot be omitted or substituted for mental means without having a material effect on the operation and structure of the embodiments described herein. The computer hardware is essential to implement the various embodiments described herein and is not merely used to perform steps expeditiously and in an efficient manner.

[0032] Fig. 2 shows an example system 200 including one or more devices 205 which may be used to control operation of multiple wells. In some examples, a device 205 may be a computational device such as a computer, server, tablet or mobile device, or other system, device or any combination thereof suitable for accomplishing the purposes described herein. In some examples, the device 205 can include one or more processor(s) 210, memories 215, and/or one or more devices/interfaces 220 necessary or desirable for input/output, communications, control and the like. The processor(s) 210 and/or other components of the device(s) 205 or system 250 may be configured to perform one or more aspects of the processes described herein.

[0033] In some examples, the device(s) 205 may be configured to receive or access data from one or more volatile or non-volatile memories 215, or external storage devices 225 directly coupled to a device 205 or accessible via one or more wired and/or wireless network(s)/communication link(s) 260. In external storage device(s) 225 can be a network storage device or may be part of or connected to a server or other device. [0034] In some examples, the device(s) 205 may be configured to receive or access data from sensors or input devices 130 in the field or infrastructure. These sensors or devices 130 may be configured for collecting or measuring well, infrastructure, operational, and/or other geological and/or physical data. In some examples, the sensor(s)/device(s) 130 can be configured to communicate the collected data to the device(s) 205 and/or storage device(s) 225 via one or more networks/links 260 or otherwise. In some examples, the sensors or devices 130 may be connected to a local computing device 250 which may be configured to receive the data from the sensors/devices 130 for local storage and/or communication to the device(s) 205 and/or storage device(s) 225. In some examples, data from sensor(s) or device(s) may be manually read from a gauge or dial, and inputted into a local computing device for communication and/or storage.

[0035] In some examples, the device(s) 205 may be configured to generate and/or transmit signals or instructions to one or more control device(s) 140 to apply desired operational inputs/conditions to the wells. These signals/instructions may, in some examples, be communicated via any single or combination of networks/links 260. In some examples, the device(s) 205 may be configured to send signals/instructions via local computing device(s) 250 connected to or otherwise in communication with the control device(s) 140. In some examples, a local computing device 250, display or other device may be configured to communicate instructions to a person for manual adjustment/control of the control device(s) 140. [0036] In some examples, a client device 260 may connect to or otherwise communicate with the device(s) 205 to gain access to the data and/or to instruct or request that the device(s) 205 perform some or all of the aspects described herein.

[0037] Fig. 3 shows a flowchart illustrating aspects of an example method 300 for controlling operation of a number of wells. In some examples, these wells can be part of the same project or may be targeting areas of the same resource. The wells may be inter-related by their sharing of one or more operational constraints such as a limited input resource shared between the wells, or a output limitation on the group of wells as a whole such as a limit on the processing capabilities of well outputs or a limit on carbon dioxide or other bi-products across the group of wells. In some instances, the wells may be part of different projects or have different well pads or processing infrastructure, but may still share one or more of the same global operating constraints. [0038] At 310, one or more processor(s) 210 and/or other aspects of device(s) 205 may be configured to generate models for each well and/or for portions (segments) of each well. The models can, in some examples, map a production rate for the respective well or well portion based on an operational input. [0039] In some embodiments, the processor(s) 210 can be configured to generate the production rate models based on historical well data. The historical well data may include data collected by input device(s) 130 for each respective well. In some examples, the historical well data may include, but is not limited to, well conditions (such as temperature, pressure and/or any other conditions as described herein or otherwise) at one or more portions of a well or at the surface. In some examples, the historical well data can include input conditions such as amounts/rates of steam (e.g. volume or volumetric rate) injection, pump speeds, water production/consumption (e.g. volume or rate), amount of carbon dioxide production which is attributable to the well's inputs, amounts/rates of solvent injection, etc. In some examples, the historical well data can include well outputs such as production rates of wells or portions of wells. In some examples, historical well data can include geological or other physical attributes of the wells or resource.

[0040] In some embodiments, in addition or as an alternative to historical well data, the processor(s) can be configured to generate production rate models based on well characteristics. In some embodiments, well characteristics may include one or more well/project/resource/formation/reservoir/drainage area identifier(s), well types, well statuses, length of time a well is in production/operation, well location (e.g. latitude, longitude, depth for surface hole, well heel, well toe, for injector and producer wells, etc.), geological or other physical attributes of the wells, etc.

[0041] In some examples, by considering locations, identifiers and/or other well characteristics, the processor(s) may generate production rate models which inherently or explicitly encompass factors such as similarities in performance between wells/formations/resources within close proximity to each other.

[0042] In some examples, by considering well characteristics such as the age or length of time a well is in production/operation, the processor(s) may generate production rate models which compensate for changes in production profiles as a SAGD well ages. [0043] In some embodiments, in addition or as an alternative to historical well data and/or well characteristics, the processor(s) can be configured to generate production rate models based on data which is interpreted, inferred or otherwise derived from historical well data (e.g. from input devices) and/or from geological or other physical well attributes. [0044] The processor(s) 210 can, in some embodiments, be configured to generate production rate models by training one or more neural networks, genetic algorithms, decision trees, or other supervised learning algorithms. In some examples, the processor(s) can be configured to use operational inputs as input variables, and production rates as outputs for the training set. In some instances, this may result in a model which inherently incorporates any well conditions and in some examples, may result in a less complex model.

[0045] In some examples, the processor(s) can be configured to additionally use well conditions and/or other attributes such as physical/geological attributes of the resource and/or age/size/shape of a steam chamber as input variables. In some examples, the processor(s) can be configured to additionally use well characteristics and/or derived data. In some instances, this may result in a more complex model (e.g. more inputs) which may be more sensitive to changes to current well conditions.

[0046] In some embodiments, generating models can include selecting inputs to the model by evaluating the influence of input variables. In some examples, this can include generating one or more models based on multiple input variables, and varying a single input while keeping all other inputs constant. By observing the effect on the output of the model as the single input changes, the processor(s) can determine a relative influence of that input. The processor(s) can be configured to repeat this process for all input variables for the model to determine which inputs have the greatest influence on the model output. In some examples, the processor(s) can be configured to generate a new, simpler model by eliminating the least influential inputs (i.e. the inputs which when varied had the smallest effect on the output) of the previous model.

[0047] In some embodiments, the processor(s) can be configured to generate models using only the 5-10 most influential inputs.

[0048] In some embodiments, the processor(s) can be configured to select the inputs used in generating a model based on the cost associated with collecting the input data. For example, repeated collection of seismic data may be costly compared to repeatedly sampling a temperature sensor positioned within a well. In some examples, the processor(s) may be configured to assign a higher weight to input parameters which are less costly to collect data. However, in some examples, when an expensive input parameter has a large influence score, the processor(s) may be configured to include the expensive input when generating a model.

[0049] In some examples, the training data set can include all historical well data. In some examples, the training data set can include a selection of the historical well data such as data collected within the last X days from when the model is generated. In some examples, the model(s) can be generated/updated on a rolling basis to reflect a model based on data from the most recent last X days. For example, in some examples, the model(s) can be generated/updated based on data from the last 30, days, 60 days, 90 days, 60 months, 1 year, etc.

[0050] In some embodiments, the processor(s) can be configured to use the number of days X of historical well data as a variable for generating a production rate. In some examples, when predicting near time behaviour (i.e. predicting a production rate for tomorrow), it may be more efficient and/or accurate to base the prediction on recent historical data. Conversely, in some examples, when predicting behaviour farther in the future, it may be more efficient and/or accurate to base the prediction on a larger history of well data. By incorporating X as a variable, the processor(s) can, in some examples, capture this flexibility when generating a model.

[0051 ] In some examples, the model may be generated based on ranges which are dependent on the age of the wells. For example, for an earlier period of a well's production cycle, the model may be based on data from a shorter period of historical well data, while for a later period of a well's production cycle, the model may be based on a larger period of historical well data. In some instances, this may accommodate a potential greater variance of a well's production at an early stage of production, and a relatively more stable production a well's production cycle. [0052] In some examples, the range of historical data used to generate the model(s) may differ for one or more variables. For example, historical temperature data for large period of time may be used, while historical oil production data for a smaller period of time may be used. Suitable periods may be dependent on the type of variable being used to generate the models. In some examples, the periods may depend on the variance of the variable and/or its effect on outputs as the well ages. In some examples, the periods may depend on whether the variable measurements are independent (e.g. temperature at any given time) or somehow related (e.g. cumulative values are based on previous measurements).

[0053] In some examples, a subset of a selection of or all of the historical data set can be used to train the model(s), while a second different subset of the historical data (not including any of the training data) can be used to verify the models. In some examples, these subsets may be randomly generated/selected from the available or selected historical data. In some embodiments, the processor(s) can be configured to generate new model(s) using different training sets if the verification of the previous model(s) does not fall within defined error tolerances. [0054] At 315, with the generated model(s), the processor(s) can be configured to assign units of operational inputs across individual wells or well portions while staying within total or global operational constraints. In some examples, the global operational constraints can be determined from data defining attributes or characteristics of a system/project's allocated resources, limitations or physical restrictions. For example, infrastructure for a group of wells may only have processing capabilities to provide a certain amount or rate of water (for creating steam) or to produce a certain amount/rate of steam. In another example, infrastructure for a group of wells may only have capabilities to process a certain rate or volume of bitumen or extracted (e.g. oil and water) mixtures. In another example, regulations or technical requirements may dictate that a group of wells only produce a certain amount/rate of carbon dioxide or other biproduct(s). In another example, there may be constraints on the amount of solvent to inject into the wells. Any of these or other factors, alone or in combination, may be used by the processor(s) as global/total operational constraints.

[0055] In some examples, the global/total operational constraints may be received or accessed from storage device(s) 225, 215 and/or memory(ies) 215 or may be received as inputs to a device 205 or system 200.

[0056] In some examples, when multiple operational constraints are defined, the processor(s) may apply weightings to determine which operational constraint may take priority over another constraint. These weightings may be received or access from a storage device or memory, or may be received as input(s) to the device 205 or system 200. [0057] The processor(s), at 315, can, in some examples, be configured to determine a distribution of operational inputs across all of the wells / well portions in the system or group of wells which provides optimal production rate(s) for the system/group of wells as a whole based on the generated models and the applied distribution method. In some embodiments, the optimal distribution of operational inputs and/or optimal resulting production rate(s) may be the best solution available based on the distribution method and the computational and/or time constraints applied to the distribution method.

[0058] In some examples, the method for determining the distribution of operational inputs may include applying genetic algorithm(s), goal seeking algorithm(s), deterministic algorithm(s), greedy algorithm(s), stochastic method(s), Newton-Raphson technique(s), Monte-Carlo method(s), and the like. In some examples, the method for determining the distribution may be iterative (as illustrated in the example in Fig. 3 and described below) and/or may be repetitive (e.g. by repeatedly assigning different inputs and determining the resulting outcomes, and selecting the best or optimal outcomes from the resulting set).

[0059] For example, the processor(s) can, in some embodiments, be configured to combine the models for the wells / well portions and the global input constraints to create one or more equations and/or models for optimization. In some examples, the processor(s) can be configured to apply Newton-Raphson method to the equation(s)/model(s) to determine the input distribution which results in a peak production rate by iteratively solving for the roots of the equation(s)/model(s).

[0060] In some examples, the processor(s) can be configured to create a genetic algorithm optimization problem using the models for the wells / well portions and the global input constraints with the fitness or objective function being the global production rate for all wells / well portions in the group.

[0061] In some examples, the processor(s) can be configured to apply a greedy algorithm wherein the processor(s) can be configured to iteratively select the current best well / well portion to assign operational inputs until a distribution is found.

[0062] In some embodiments, the above optimizations may be repeated any number of times or as many times as possible within a defined time/process limitation to try to ensure that the optimum distribution found is closer to a global peak production rate rather than a local maximum. [0063] In other embodiments, any other suitable optimization method or combination of methods may be used.

[0064] In some embodiments, at 330, the processor(s) can be configured to assign a unit of an operational input to a well or portion of a well which would result in the greatest incremental production rate for the system or group of wells as a whole.

[0065] In some examples, one or more of the operational inputs may be the same as one or more of the global operational constraints. For example, a global constraint may include a limit on the amount of steam available to a group of wells, while an operational input may include the amount of steam to provide to a particular well or portion of a well. [0066] In some examples, one or more operational inputs may be different than one or more of the global operational constraints. For example, a global constraint may include a limit on the amount of carbon dioxide which can be produced across a group of wells, while an operational input may include the amount of steam to provide to a particular well or well portion. However, the assigning of a unit of the operational input (i.e. unit rate of steam) will create carbon dioxide when producing the steam, delivering the steam and/or pumping at least a portion of the water from the producer well after the steam condenses. In some embodiments, the processor(s) are configured to determine the incremental effect of assigning the unit of operational input on the global operational constraint.

[0067] In some examples, the processor(s) can be configured to determine the well or well portion which would result in the greatest incremental production by applying the unit of operational input to each well/well portion model and comparing the resulting incremental production rates.

[0068] After assigning the unit of operational input, the processor(s) can be configured to determine the incremental effect of the operational input on the operational constraints. [0069] At 340, this process of assigning units of operational inputs to the next well / well potion having the greatest incremental production rate and determining the effect on the operational constraints is repeated until one or more of the global operational constraints are reached. [0070] In some situations, data associated with wells or well portions may indicate that minimum or maximum operational inputs are required to maintain a well or well portion. For example, if the temperature or pressure in a well or well portion gets too low, a steam chamber or other physical aspects associated with a well may be permanently damaged which may affect all future production from the well. In another example, regulatory requirements and/or the location of a well may dictate for example maximum pressure limits to avoid breaches, etc. Therefore, minimum and/or maximum operational inputs can be associated with a well as defined by a database or input in order to maintain a well's attributes (such as temperature or pressure). [0071 ] Optionally, at 320, before incrementally assigning units of operational inputs at 330, the processor(s) may be configured to assign units of operational inputs to meet minimum or maximum operational conditions.

[0072] Once one or more of the global operational constraints are reached, or once an optimal distribution has been determined, at 350, the processor(s) can be configured to apply the total operational inputs to each well or well portion. In some examples, applying the operational inputs can include generating signals or instructions for controlling or otherwise modifying or maintaining the operation of one or more control devices 140. In some examples, the signals or instructions can be communication via the network(s)/link(s) 260. In some examples, the signals or instructions can be audibly or visually communicated (e.g. on a display, printer output, speaker, etc.) for manual configuration or adjustment of the control devices 140.

[0073] In some embodiments, before applying the operational inputs, the processor(s) can be configured to perform a sanity check on the determined optimal distribution. In some examples, the processor(s) may be configured to compare the determined predicted optimal production rate with a previous distribution's predicted or actual measured production rate. If the predicted rate is greater than or less than the previous production rate by a defined warning threshold (e.g. extremely large or extremely small different), the processor(s) may be configured to generate a warning message and/or to regenerate the model(s) and/or distribution.

[0074] In some embodiments, before applying the operational inputs, the processor(s) can be configured to verify that the operational inputs for the determined optimal distribution do not violate the volumetrics or other constraints on the wells and/or infrastructure. [0075] At 360, the processor(s) can be configured to receive well production and/or well condition data from the input device(s) 130 associated with the application of the operational inputs.

[0076] In some examples, the new well production data, the well condition data and/or the operational inputs from 360 can be incorporated to regenerate or update the model(s) for the wells / well portions.

[0077] In some examples, the processor(s) can be configured to verify the expected production results based on the models and the applied operational inputs against the actual received production data. In some examples, the regeneration or updating of the model(s) is only triggered when the expected production results vary from the actual production data by a defined error threshold.

[0078] In some examples, the regeneration or updating of the model(s) is automatically triggered periodically (e.g. every X days or weeks).

[0079] In any of the above examples, or otherwise, in some embodiments, any time the models are regenerated or global constraints change, the iterative assignment of operational inputs and their application to the wells/well portions is repeated.

[0080] In some examples, the influence of different variables may change as the age of the well (e.g. time a well has been in operation/production). For example, in some instances, a young well's production rate may be highly sensitive to the distance between the injector and producer wells, and to the rate of steam injection. However, in some instances, the well's production rate may become less sensitive to these parameters as the well ages. For example, in some instances, an older well's production rate may more greatly influenced by the geological characteristics of the formation or by the historic production rates of surrounding wells. In some examples, to account for this or otherwise, the processor(s) may be configured to automatically trigger the regeneration of model(s) after a well reaches defined age threshold(s).

[0081] In some examples, the processor(s) can be configured to receive input(s) from one or more input devices 130 or user input devices indicating that a malfunction, shutdown or maintenance event for one or more of the wells has or is scheduled to occur. Upon receipt of an input, the processor(s) can be configured to regenerate the model(s) for the affected well(s) / well portion(s) and reassign and apply the operational inputs based on the regenerated model(s).

[0082] In some examples, for future scheduled events, the processor(s) may be configured to regenerate model(s) in incremental stages in order to gradually slow down or turn off a well. In some scenarios, this may help prevent permanent damage to a well cause, for example, by a rapid change in temperature or pressure.

[0083] Although the embodiments have been described in detail, it should be understood that various changes, substitutions and alterations can be made herein without departing from the scope as defined by the appended claims. [0084] Moreover, the scope of the present application is not intended to be limited to the particular embodiments of the process, machine, manufacture, composition of matter, means, methods and steps described in the specification. As one of ordinary skill in the art will readily appreciate from the disclosure of the present invention, processes, machines, manufacture, compositions of matter, means, methods, or steps, presently existing or later to be developed, that perform substantially the same function or achieve substantially the same result as the corresponding embodiments described herein may be utilized. Accordingly, the appended claims are intended to include within their scope such processes, machines, manufacture, compositions of matter, means, methods, or steps

[0085] As can be understood, the examples described above and illustrated are intended to be exemplary only. The scope is indicated by the appended claims.