Patent Application
20190186237
The subject matter described herein relates to methods and control systems that distribute allocated quantities of steam to multiple wells in steam assisted gravity drainage (SAGD) resource production applications.
SAGD is an enhanced oil recovery technique for producing heavy crude oil and bitumen which can be used in oil sands projects, for example. The technique involves drilling two parallel horizontal wells. An upper well (that is closer to the ground surface than a lower well) injects steam into a geological formation to form a steam chamber, and the lower well collects heated crude oil or bitumen, as well as formation water and water from condensation of the steam that flows out of the steam chamber. The steam reduces the viscosity of the bitumen allowing the bitumen to flow with the water and other minerals and chemicals to the lower well. The condensed mixture of oil and water (and other components) in the lower well are pumped to the surface where the oil is separated from the mixture. There may be approximately three to five times as much water used in SAGD resource production than the amount of oil produced.
The cost of heating the water in the steam generator to produce the steam for SAGD production is a large factor in the operating cost of SAGD operations. Efficient steam utilization is important for successful SAGD operations, such that efficient allocation of the steam to the multiple wells in a well pad can provide enhanced resource recovery compared to less efficient allocation of the steam to the wells. But, it is difficult to fully characterize and understand the subterranean reservoir in order to determine how to efficiently allocate the steam among the wells in the well pad. Some methods may use limited observations and/or intuition alone to provide insights for controlling the allocation of the limited steam resources, which are not that successful at increasing steam utilization or efficiency and/or increasing oil production relative to equal or random allocation of steam to the wells. Other methods attempt to determine an “optimal” outcome (e.g., oil production, steam-to-oil ratio, or steam utilization) using computer modeling methods, but the models are complex and involve relatively high number of variables. As a result of the complexity, the modeling is very time intensive, often requiring at least six months to multiple years to generate an outcome scheme that can be employed in the field. The long time delay is undesirable as it delays efficient use of the land until the modeling is completed. Furthermore, the geology of the subterranean reservoir may change in the intervening time period while the computers are running the model, such that the data in the model may be stale and outdated. As a result of the changing conditions over time, the outcome generated by the model may not be an “optimal” outcome for the current conditions of the subterranean reservoir.
In an embodiment, a system (e.g., a control system for steam assisted gravity drainage (SAGD) resource production) is provided that includes a steam distributor and one or more processors. The steam distributor is configured to distribute steam received from a steam generator to multiple injection wells in a well pad for SAGD resource production. The one or more processors are configured to control the steam distributor to distribute the steam to the injection wells according to allocated quantities of steam designated in a resultant scheme. The resultant scheme includes values representing multiple parameters for the SAGD resource production. The parameters include the allocated quantities of steam to the injection wells, pressures within the injection wells and within multiple production wells associated with the injection wells, and time periods that the steam is directed into the injection wells. The one or more processors are configured to determine the resultant scheme by performing multiple iterations of a surrogate evaluation process until a stop criterion is met. The one or more processors perform a first iteration of the surrogate evaluation process by designing a first sample scheme and then evaluating the first sample scheme using a reservoir simulation model to provide a predicted resource output associated with the first sample scheme. The one or more processors are configured to sequentially repeat the surrogate evaluation process for one or more additional iterations using sample schemes having different values of the parameters than preceding sample schemes previously evaluated in the reservoir simulation model. The values of the parameters in the sample scheme of each of the additional iterations are selected based on the predicted resource outputs associated with the preceding sample schemes. The one or more processors are configured to identify the resultant scheme as the sample scheme of a final iteration prior to the stop criterion being met.
In another embodiment, a method (e.g., for performing steam assisted gravity drainage (SAGD) resource production) is provided that includes evaluating plural schemes for allocating steam to well pairs in a well pad for SAGD resource production. The plural schemes include an initial set of schemes. The schemes are evaluated using a reservoir simulation model. Each scheme includes values representing multiple parameters for the SAGD resource production. The parameters include one or more of quantities of steam allocated to the well pairs, pressures within the well pairs, and time periods that the steam is directed into the well pairs. Evaluation of each of the schemes using the reservoir simulation model provides a predicted resource output associated with the corresponding scheme. The method also includes performing a surrogate evaluation process for a first iteration. The surrogate evaluation process includes designing a first sample scheme that has different values of the parameters for SAGD resource production than the schemes in the initial set. The values of the parameters in the first sample scheme are selected based on the predicted resource outputs associated with the schemes in the initial set. The surrogate evaluation process includes evaluating the first sample scheme using the reservoir simulation model to provide a predicted resource output associated with the first sample scheme. The method further includes sequentially repeating the surrogate evaluation process for one or more additional iterations using sample schemes having different values of the parameters than preceding schemes that include the first sample scheme and the schemes in the initial set. The values of the parameters in the sample scheme of each of the additional iterations are selected based on the predicted resource outputs associated with the preceding schemes. The surrogate evaluation process is sequentially repeated until a stop criterion is met. The method includes identifying a resultant scheme for controlling the SAGD resource production according to the values of the parameters contained in the resultant scheme. The resultant scheme is identified as the sample scheme of a final iteration prior to the stop criterion being met.
In another embodiment, a system (e.g., a control system for steam assisted gravity drainage (SAGD) resource production) is provided that includes a controller configured to evaluate different parameters for allocating steam to well pairs in a well pad for steam assisted gravity drainage (SAGD) resource production. The controller is configured to determine expected resource outputs from the well pad based on one or more steam introduction parameters. The steam introduction parameters include one or more of different quantities of steam directed into the well pairs, different pressures controlled within the well pairs, or different time periods that the steam is directed into the well pairs. The controller is also configured to control introduction of the steam into the well pairs using at least one of the steam introduction parameters based on the expected resource outputs.
The inventive subject matter described herein will be better understood from reading the following description of non-limiting embodiments, with reference to the attached drawings, wherein below:
Reference will be made below in detail to example embodiments of the inventive subject matter, examples of which are illustrated in the accompanying drawings. Wherever possible, the same reference numerals used throughout the drawings refer to the same or like parts.
The methods and systems described herein can be used to generate a scheme for controlling allocation of steam to various wells in a well pad in a SAGD production in order to increase a production outcome of a resource and/or a steam utilization efficiency relative to operation of the well pad according to a different scheme, such as a scheme determined based on intuition and/or limited observations. The scheme designates values for parameters associated with the SAGD resource production, such as an allocated amount of steam to each injection well in the well pad, a time period for injecting steam into each of the injection wells, an injection pressure at which steam is injected into each of the injection wells, and the like. The methods and systems described herein are configured to generate multiple schemes and evaluate the schemes in a reservoir simulation model sequentially, such that the details or parameters described in each subsequent scheme are chosen based on results from the evaluations of the previous schemes in the reservoir simulation model.
In one or more embodiments described herein, the methods and systems identify principal parameters in the SAGD production and utilize meta-modeling (including smart sampling and surrogate modeling) to generate the scheme for controlling the distribution of the steam and other parameters in the SAGD production. The meta-modeling uses the design space defined by the identified principal parameters to intelligently and sequentially extract subsequent schemes to evaluate in the reservoir simulation model. Each extracted scheme is exercised or evaluated in the reservoir simulation model to output results that are utilized to develop a surrogate model. Over a sequential process, the surrogate model is updated to reflect output results from previous simulations. Using the sequential surrogate model, subsequent schemes may be developed that gradually converge towards a resultant scheme that is configured to increase the production outcome (e.g., cumulative oil production, steam-to-oil ration, steam utilization efficiency, etc.) relative to operation of the well pad according to a different scheme. The surrogate model allows for a significant reduction in the number of investigations or evaluations that are performed on the reservoir simulation model compared to scheme selection techniques that use space-filling designs and large sample sets to capture the complexity of the response surface. For example, instead of hundreds of evaluations, the surrogate model is able to reach a resultant scheme with less than 20 evaluations. Therefore, the surrogate modeling described herein can reduce the calculation period by five times or more, such that a solution can be generated in a matter of weeks instead of months or years.
One or more technical effects of the methods and systems disclosed herein includes the ability to design a field-specific SAGD production operation that increases economic value for outcomes such as oil production, steam utilization, and steam-to-oil ratio relative to SAGD production operations that do not utilize the described surrogate modeling. Another technical effect is the ability to generate a resultant scheme or scenario for designing the SAGD production parameters in a more time-effective manner, in a matter of days or weeks instead of months or years, which allows for earlier commencement of SAGD production in a given subterranean reservoir and more timely and accurate data relative to waiting months or years to generate a resultant scheme. Yet another technical effect is the flexibility of being able to use the system and method on a field-specific basis, which allows customized field-specific strategies for steam allocation. Still another technical effect is the possibility and feasibility for operators to monitor and track resource (e.g., oil) recovery and injection parameters in a timely manner within a confidence interval using the surrogate meta-modeling.
The well pad 102 shown in
The steam from the generator 122 is directed to a steam distributor 124 which distributes the steam received from the generator 122 to the multiple well pairs 104. For example, the steam distributor 124 receives a single input stream and distributes the steam to multiple output streams that are each directed to a different well pair 104 in the well pad 102. The steam distributor 124 may include a plurality of insulated steam distributor tubes, valves, conduits, and/or the like for controlling an amount of steam directed to each of the well pairs 104. The steam distributor 124 may also condition the steam, such as by controlling a pressure and a humidity of the steam directed to the well pairs 104. The steam distributor 124 is able to unevenly distribute the steam among the well pairs 104 such that some well pairs 104 receive more steam (e.g., a higher steam flow rate and/or receive steam for longer durations) than other well pairs 104 in the well pad 102. The steam distributor 124 is also able to reallocate the steam among the well pairs 104 by changing the corresponding amounts or rates of steam directed to at least some of the well pairs 104. Therefore, the steam distributor 124 may supply steam at a first flow rate to one of the well pairs 104 for a first time period, and then may supply steam at a different, second flow rate to the same well pair 104 for a subsequent, second time period. In an embodiment, the steam distributor 124 is a series of insulated pipes or tubes that define four steam conduits extending to the four well pairs 104. Each conduit includes one or more electrically-operated valves and/or pumps for controlling a flow rate and pressure of steam flowing through the respective conduit.
In an embodiment, the steam distributor 124 is controlled by a controller 126. For example, the controller 126 may control the valves and/or pumps of the steam distributor 124 to control the distribution of steam to each of the well pairs 104. The controller 126 is communicatively connected to the steam distributor 124 such that the controller 126 conveys electromagnetic control signals to the steam distributor 124 to control the operation of the steam distributor 124. For example, the controller 126 may convey electrical control signals directly to the valves and/or pumps of the steam distributor 126, or may convey the signals to local control circuitry on the steam distributor 126. The controller 126 transmits the control signals to the steam distributor 124 via a conductive path along one or more wires or wirelessly using a communication circuit (e.g., a transceiver and associated circuitry). The controller 126 represents hardware circuitry that includes, represents, and/or is connected with one or more processors 128 (e.g., microprocessors, field programmable gate arrays, integrated circuits, or other electronic logic-based devices). The controller 126 may include and/or be communicatively connected with one or more digital memories, such as computer hard drives, computer servers, removable hard drives, etc. Although shown as a separate, standalone device, the controller 126 optionally may be mounted on the distributor 124 or on another component of the control system 100.
The steam distributor 124 directs steam to the injection wells 106 of the well pairs 104. The steam flows through the injection wells 106 and is injected into the subterranean reservoir 110. The heat from the steam reduces the viscosity of the oil-based resource (e.g., heavy crude oil or bitumen), which allows the resource to flow downwards via gravity into the lower production well 108. The injected steam forms a steam chamber 130, as shown in
A mixture of the oil-based resource, condensed water, and other solvents and minerals are collected in the production wells 108 of the well pad 102 and recovered to the ground surface 116 by corresponding pumps 132. The pumps 132 may be progressive cavity pumps, electric submersible pumps, or the like, that are configured to propel high-viscosity fluids and suspended solids. Although illustrated in
The mixture of the oil-based resource, the condensed water, and other minerals and solvents is directed from the production wells 108 to an aggregator 134. The aggregator 134 may be a container, such as a tank. The mixture is fed from the aggregator 134 to a separator 136. The separator 136 is configured to remove the oil-based resource from the remaining components in the mixture. The separator 136 may include filters, separation vessels, gas flotation vessels, or the like. The separated oil-based resource (e.g., bitumen) is directed to a product tank 138, while a remaining oil/water mixture referred to as process water is directed to a water treatment system 140. The water treatment system 140 may include a second oil-water separation process, such as via filtering, chemical processing, gas flotation or other evaporative processes, or the like. Bitumen recovered from the water treatment system 140 may be directed to the product tank 138. Some of the treated water may be recycled to the steam generator 122 for generating more steam, and another portion of the treated water may be purged 142 and released into a body of water or pumped underground.
Some parameters are associated with the geology of the subterranean reservoir and other parameters are associated with operational control parameters. The geological parameters may include a porosity of the subterranean reservoir, a permeability of the subterranean reservoir, reservoir initial saturation condition, the type of rocks and minerals in the subterranean reservoir, and the like. The operational control parameters may include the locations of the well pairs, the distances between the well pairs, the spacing between the wells in each pair, the number of well pairs, an amount of steam allocated to each well pair, designated reallocations of the amounts of steam distributed to the well pairs over time, a pressure in the production well of each well pair, a degree of sub-cool in the production wells, a steam flow rate or injection rate into each injection well, time periods that steam is injected into each of the injection wells, a temperature of the steam, a quality of the steam, and/or the like.
Each of the schemes designates values for the parameters for SAGD production. For example, a first initial scheme may allocate 100 cubic meters (m3) per day of steam to the injection well of a first well pair, 150 m3/day of steam to the injection well of a second well pair, 200 m3/day of steam to the injection well of a third well pair, and 225 m3/day of steam to the injection well of a fourth well pair. A second initial scheme may have a different allocation of steam among the well pairs, such that more steam is distributed to the injection well of the first well pair than the injection well of the fourth well pair. The steam capacity or total available amount of steam to be distributed among the well pairs per day may be a fixed quantity that is treated as a hard constrain. For example, although the different initial schemes may have different allocations of steam to the well pairs, the cumulative amount of steam distributed to the well pairs in each scheme does not exceed the steam capacity. The steam capacity may be dependent on the water source 120, the capacity of the steam generator 122, a law or regulation, or may be selected based on economic principles. In addition to steam allocation, the initial schemes may also differ with respect to the values for other parameters, such as the pressures within the injection wells and production wells of the well pad.
The production of the oil-based resource (e.g., bitumen) and the steam utilization rate are affected by the parameters. Due to the complexity and number of the parameters, it is difficult to compare the schemes to determine which schemes are better than other schemes at increasing the production of bitumen relative to the usage of steam. Therefore, at 403, a reservoir simulation model (e.g., the reservoir simulation model 144 shown in
The reservoir simulation model is configured to run simulations for each of the schemes in the initial set of schemes. The simulations are designed to model the effects of performing the SAGD operation according to each scheme over a period of time, such as five or ten years. The reservoir simulation model may model both the field-specific environment and the operating parameters according to the scheme. For example, in an embodiment, the reservoir simulation model incorporates a static geological model and fluid dynamic properties. The static geological model represents the geometry and character of the porous media of the reservoir. The fluid dynamic properties describe the nature of the various fluid phases present in the reservoir. The static geological model and fluid dynamic properties may be stored in a digital memory that is accessed by one or more processors. The reservoir simulation model includes a history-match aspect in which the current performance parameters are compared to historical performance and adjustments are made to reasonably represent reality. For example, the reservoir simulation model can compare the parameters/input of a new geological environment with parameters of previous geological environments with known outcomes (e.g., known amounts of resource extraction for various inputs of steam, permeability and porosity of the subterrain, and the like). Based on similarities and/or differences with the previous environments, the simulation can estimate production from the new environment.
At 404, the schemes in the initial set of schemes are evaluated using the reservoir simulation model. Due to the number and complexity of the parameters, each simulation using the reservoir simulation model may be time intensive, requiring approximately 10-24 hours or longer to complete. The initial set of schemes may include less than 10 schemes, such as 3-5 schemes. The values for the parameters of the schemes may be selected using a space-filling design function, such as Latin Hypercube.
Each evaluation of a scheme using the reservoir simulation model may provide an objective production outcome. The objective production outcome may be provided in various forms or formats, but generally indicates a cumulative amount of the oil-based resource produced per a cumulative amount of steam utilized in a given amount of time according to the specifications of the respective scheme that is evaluated. For example, the objective production outcome may be provided as a resource output (e.g., cumulative amount of produced oil, bitumen, or the like per a given time period), a steam input (e.g., cumulative amount of steam used in a given time period), a steam utilization or efficiency value, a steam-to-oil ratio (SOR), and/or the like. The SOR indicates how many units steam is used to produce one unit of oil. The objective production outcome may be combined with economic data (e.g., oil prices, steam prices, interest rates, inflation rates, etc.) to provide the production outcome in terms of net present value, net operating cost, or the like. In general, schemes that provide lower SOR, greater resource output, and/or lower steam input are more desirable than schemes that provide higher SOR, lower resource output, and greater stem input. As used herein, the objective production outcomes are typically referred to in terms of predicted steam-to-oil ratio (SOR) and predicted resource output, but it is recognized that the SOR and resource output can be converted into the other outcomes described above, such as steam input or net present value, by simple calculations. For example, the SOR can be calculated by dividing the steam input by the resource output.
At 406, a surrogate model (e.g., the surrogate model 146 shown in
The surrogate model is developed by identifying principal parameters that have a greater effect on the resulting outcome of the evaluation (e.g., the predicted resource output and/or SOR) than other parameters in the SAGD production operation. For example, the parameters may be ranked or weighted in order of relative influence on the outcome. The principal parameters may include steam allocation (e.g., the quantity of steam allocated to each well pair), pressures within the well pairs, and time periods that steam is directed into the well pairs at a current steam allocation. The steam allocation may include or represent a steam injection rate, such that the allocation may be a quantity of steam distributed to each well pair per day. The pressures within the well pairs may correspond to the pressure within the production well, as controlled in part by the pumps. The time periods that steam is directed into the well pairs may designate pre-soak times before the production well begins collecting and pumping the oil-based resource to the surface, times that the injection wells receive steam, and/or times that the steam allocation to each of the well pairs may be changed according to the scheme. The principal parameters optionally may also include other parameters, such as steam temperature, steam quality, well spacing, distance between well pairs, degree of sub-cool, and/or the like.
The surrogate model may be a regression model that is designed to emulate the full-scale reservoir simulation model. The surrogate model is based on the identified principal parameters contained in the schemes of the initial set and is also based on the outcomes of the evaluations of the schemes, such as the resource output. The surrogate model represents the responses of the principal parameters for each evaluation in the reservoir simulation model. For example, the surrogate model may match and/or compare the results from the evaluations of the schemes in the initial set, and may narrow spectrum bands to be used in selecting the values of the parameters for a subsequent scheme.
Optionally, a Gaussian regression may be used to build the surrogate model. The Gaussian regression characterizes an unknown function in terms of a Gaussian distribution over functions fully specified by a mean value and a correlation structure on the sampling input space, expressed as a kernel function. The mean value models the expected value of the underlying function being modeled and the kernel models the properties of the input-output response surface like smoothness. Starting with an infinite set of functions, the process works by narrowing down the function-set by conditioning it on actual values of the current set of sample evaluations using Bayesian estimation.
At 408, a determination is made whether a stop criterion is met. The stop criterion may be based on a designated number of evaluations using the reservoir simulation model, a designated time limit, or a calculated amount of variance in the outcome of the evaluations that is less than a designated threshold value (e.g., which indicates that the results of the simulations are converging). The determination at 408 is discussed in more detail below. The stop criterion is not met on a first run through the method 400. The method 400 continues to 410 and a subsequent scheme (e.g., a first sample scheme) is designed based on the surrogate model.
The first sample scheme has different values of the parameters for the SAGD production than the schemes in the initial set. The values of the parameters in the first sample scheme are selected using the surrogate model. The first sample scheme is designed based on the results of the evaluations of the initial set of schemes, such as the resource output for the initial set of schemes, because the results of the evaluations are used to develop the surrogate model. In an embodiment, the surrogate model selects the values for the first sample scheme based in part on the underlying problem for which the samples are being extracted, which is increasing the resource output and/or reducing the SOR. For example, the surrogate model can represent a set of functions plotted in a sample space. The new sample schemes may be selected in regions of the sample space where variance or uncertainty in the functions is relatively high and where a mean value of the Gaussian regression is low, representing locations where the SOR may be low. Therefore, the first sample scheme may be designed in an area of the sample space that balances both relatively high variance and relative proximity to a desired result (e.g., a low SOR). In another embodiment in which the desired outcome is increased resource output, the new sample schemes may be selected in regions of the sample space where a mean value of the Gaussian regression is high, representing locations where the bitumen production levels may be greater than other locations having lower mean values.
At 412, the first sample scheme is evaluated using the reservoir simulation model. The evaluation of the first sample scheme is similar to the evaluations of the schemes in the initial set as described in step 404, and so will not be described in detail. The evaluation of the first sample scheme provides a predicted resource output associated with the first sample scheme. Once the first sample scheme is evaluated, flow of the method 400 returns to 406 and the surrogate model is updated using the results, such as the predicted resource output, of the first sample scheme.
The series of steps 406, 410, and 412 may be referred to herein as a surrogate evaluation process. The surrogate evaluation process begins with the development or updating of the surrogate model, then proceeds to design of a subsequent sample scheme using the surrogate model, and ends with the evaluation of the subsequent sample scheme using the reservoir simulation model. In an embodiment, the method 400 includes performing the surrogate evaluation process for multiple iterations until the stop criterion is met at 408. For example, the surrogate evaluation process is repeated sequentially for multiple iterations. The first iteration includes the design and evaluation of the first sample scheme as described above. Upon returning to 406 to update the surrogate model, a second iteration of the surrogate evaluation process begins. If the stop criterion is not met at 408, then flow continues to 410 during the second iteration. At 410, a subsequent sample scheme (e.g., a second sample, scheme) is designed based on the surrogate model, which has been updated after the first iteration. The second sample scheme has different values for the parameters for SAGD production than all preceding schemes including the first sample scheme and the schemes in the initial set. Therefore, the values for all or at least some of the parameters are varied for each subsequent scheme. The second sample scheme is designed based on the predicted resource outputs associated with the preceding schemes, because the predicted resource outputs associated with the preceding schemes are used to construct the surrogate model that designs the subsequent schemes. The surrogate evaluation processes are performed iteratively such that the second sample scheme is not designed until after the first sample scheme is evaluated using the reservoir simulation model and the results of which are used to update the surrogate model. After the second sample scheme is designed, the second sample scheme is evaluated using the reservoir simulation model at 412. The culmination of the evaluation indicates the end of the second iteration of the surrogate evaluation process. The results of the evaluation are used at 406 to update the surrogate model again, beginning a third iteration of the surrogate evaluation process.
In an embodiment, after several iterations of the surrogate evaluation process, the results of the evaluations using the reservoir simulation model may begin to converge.
Referring now back to
Responsive to determining that the stop criterion has been met at 408, the flow of the method 400 continues to 414 and a resultant scheme is identified. The resultant scheme is identified as the latest or most recent sample scheme that is designed prior to the stop criterion being met. For example, referring to the graph 500 in
At 418, allocation of steam to the well pairs for SAGD production is controlled based on the resultant scheme. For example, the well pad is operated for SAGD production based on the values of the parameters designated in the resultant scheme. The values include the steam allocation to each of the well pairs, and also include other parameters such as designated pressures in the production wells and time periods that the injection wells receive steam. The resultant scheme is used as a plan or guideline for operation of the well pad. As shown in the graph 500 of
Optionally, during the SAGD production operation according to the parameters defined in the resultant scheme, the method 400 may determine at 420 whether or not to revise the scheme. During operation in which the steam is distributed to the well pairs, various parameters may be monitored and reviewed to determine whether the scheme should be updated. One parameter may be steam capacity. For example, if there is more steam available than expected, the scheme may be revised in order to distribute more steam to the well pairs to increase the cumulative resource output. Similarly, if there is less steam available than expected, the scheme may be revised in order to reallocate the steam among the well pairs to reduce the amount of steam utilized. The determination may be made by the controller 126 (shown in
If, on the other hand, it is determined that a revised scheme is warranted, then flow of the method 400 may continue to 422 and a new set of schemes is designed that incorporate the updated values of the monitored parameters. The new set of schemes are designed based on the resultant scheme and updated parameter data. For example, if the steam capacity has changed by an amount that is more than expected, then the updated steam capacity is used to generate one or more new schemes in the new set of schemes. The new schemes are evaluated by the surrogate model at 406 and used to determine an updated resultant scheme when the method 400 returns to 414. Therefore, it is recognized that portions of the method 400 may be repeated prior to or after starting the SAGD production in order for the resultant scheme to be based on relevant parameter data. The revision steps may be performed automatically such that the resultant scheme that is used to control SAGD production parameters may be updated and modified automatically one or more times during the course of the SAGD resource production.
The method 400 is configured to provide a resultant scheme for controlling operations of a SAGD production system with significantly reduced lapse time (e.g., the total time spent to generate the resultant scheme) relative to designing the schemes based on space-filling or other techniques. For example, using space-filling sample selection, hundreds of evaluations using the reservoir simulation model may be required before convergence is detected and/or a resultant scheme is determined. The lapse time for evaluating 200 sample schemes could take between six months and two years, depending on the time required for each evaluation. The methods and systems described herein are configured to reduce the lapse time by reducing the total number of evaluations performed using the reservoir simulation model to, for example, less than 20 total evaluations before determining a resultant scheme that is used to control the SAGD production.
In an embodiment, a system is provided that includes a steam distributor and one or more processors. The steam distributor is configured to distribute steam received from a steam generator to multiple injection wells in a well pad for steam assisted gravity drainage (SAGD) resource production. The one or more processors are configured to control the steam distributor to distribute the steam to the injection wells according to allocated quantities of steam designated in a resultant scheme. The resultant scheme includes values representing multiple parameters for the SAGD resource production. The parameters include the allocated quantities of steam to the injection wells, pressures within the injection wells and within multiple production wells associated with the injection wells, and time periods that the steam is directed into the injection wells. The one or more processors are configured to determine the resultant scheme by performing multiple iterations of a surrogate evaluation process until a stop criterion is met. The one or more processors perform a first iteration of the surrogate evaluation process by designing a first sample scheme and then evaluating the first sample scheme using a reservoir simulation model to provide a predicted resource output associated with the first sample scheme. The one or more processors are configured to sequentially repeat the surrogate evaluation process for one or more additional iterations using sample schemes having different values of the parameters than preceding sample schemes previously evaluated in the reservoir simulation model. The values of the parameters in the sample scheme of each of the additional iterations are selected based on the predicted resource outputs associated with the preceding sample schemes. The one or more processors are configured to identify the resultant scheme as the sample scheme of a final iteration prior to the stop criterion being met.
Optionally, the stop criterion is based on at least one of a number of iterations, expiration of a designated time period, or a measured variance in the predicted resource outputs associated with the sample schemes in a group of recent iterations being less than a designated threshold value. Optionally, the stop criterion is met responsive to the measured variance in the predicted resource outputs associated with the sample schemes in the group of recent iterations being less than the designated threshold value. The group includes the sample schemes from a previous three iterations before the stop criterion is met. The designated threshold value is no greater than 5%.
Optionally, the one or more processors are configured to design the sample schemes of the additional iterations such that all of the designated quantities of steam allocated to the injection wells in one sample scheme are different than all of the designated quantities of steam allocated to the injection wells in the preceding sample schemes.
Optionally, the one or more processors are configured to control the steam distributor to vary the quantities of steam allocated to the injection wells over time according to the resultant scheme.
Optionally, a predicted resource output associated with the resultant scheme that is provided by the reservoir simulation model is greater than the predicted resource outputs associated with the preceding sample schemes.
Optionally, the one or more processors are configured to perform the surrogate evaluation process for no more than 20 iterations before the stop criterion is met.
Optionally, the one or more processors are configured to use the reservoir simulation model to provide the predicted resource output and at least one of a predicted steam input or a predicted steam-to-resource ratio associated with the first sample scheme. The one or more processors are configured to select the values of the parameters in the sample scheme of each of the additional iterations based on the predicted resource outputs and the at least one of the predicted steam input or the predicted steam-to-resource ratio associated with the preceding sample schemes.
Optionally, the one or more processors are configured to design the sample schemes of the additional iterations such that the allocated quantities of steam to the injection wells in the well pad do not exceed a designated steam capacity and the pressures in the injection wells and the production wells are maintained within respective operating pressure ranges.
Optionally, the multiple parameters for SAGD resource production in each sample scheme further include one or more of number of injection wells, relative distances between the injection wells, relative spacing between each injection well and each associated production well, locations of the injection wells and the production wells, steam injection rates into the injection wells, steam quality provided by the steam generator, steam temperature provided by the steam generator, degree of sub-cool in the production wells, porosity of a subterranean reservoir, and permeability of the subterranean reservoir.
Optionally, the injection wells and the production wells define multiple well pairs in the well pad. Each well pair includes one injection well and one production well. The well pad includes between two and ten well pairs. The multiple parameters for SAGD resource production in each sample scheme further include a number of well pairs in the well pad.
Optionally, the system includes pumps disposed within the production wells that are associated with the injection wells. The one or more processors are configured to control operation of the pumps such that the pressures within the production wells approximately match the pressures designated in the resultant scheme.
Optionally, each injection well is associated with a corresponding production well to define a well pair. The injection and production wells in each well pair extend relatively horizontal and parallel to each other along a length through a subterranean reservoir. The injection well is disposed vertically between the production well and a ground surface above the subterranean reservoir. The injection well is configured to receive the steam from the steam distributor and inject the steam into the subterranean reservoir. The production well is configured to collect an oil-based resource from the subterranean reservoir that is heated by the steam.
In another embodiment, a method is provided that includes evaluating plural schemes for allocating steam to well pairs in a well pad for steam assisted gravity drainage (SAGD) resource production. The plural schemes include an initial set of schemes. The schemes are evaluated using a reservoir simulation model. Each scheme includes values representing multiple parameters for the SAGD resource production. The parameters include one or more of quantities of steam allocated to the well pairs, pressures within the well pairs, and time periods that the steam is directed into the well pairs. Evaluation of each of the schemes using the reservoir simulation model provides a predicted resource output associated with the corresponding scheme. The method also includes performing a surrogate evaluation process for a first iteration. The surrogate evaluation process includes designing a first sample scheme that has different values of the parameters for SAGD resource production than the schemes in the initial set. The values of the parameters in the first sample scheme are selected based on the predicted resource outputs associated with the schemes in the initial set. The surrogate evaluation process includes evaluating the first sample scheme using the reservoir simulation model to provide a predicted resource output associated with the first sample scheme. The method further includes sequentially repeating the surrogate evaluation process for one or more additional iterations using sample schemes having different values of the parameters than preceding schemes that include the first sample scheme and the schemes in the initial set. The values of the parameters in the sample scheme of each of the additional iterations are selected based on the predicted resource outputs associated with the preceding schemes. The surrogate evaluation process is sequentially repeated until a stop criterion is met. The method includes identifying a resultant scheme for controlling the SAGD resource production according to the values of the parameters contained in the resultant scheme. The resultant scheme is identified as the sample scheme of a final iteration prior to the stop criterion being met.
Optionally, the method further includes controlling an allocation of steam to the well pairs based on the designated quantities of steam allocated to the well pairs described in the resultant scheme.
Optionally, the values of the parameters of the sample schemes in the one or more additional iterations are selected based on the predicted resource outputs associated with the preceding schemes to increase the predicted resource output of the sample schemes relative to the predicted resource outputs in the preceding schemes.
Optionally, the predicted resource output for each scheme indicates an amount of oil-based resource that is cumulatively produced by the well pairs of the well pad responsive to injecting steam into the well pairs of the well pad according to the values of the parameters designated in the corresponding scheme.
Optionally, the values of the parameters in the first sample scheme and the sample schemes in the additional iterations are selected using a Gaussian regression.
Optionally, the values of the parameters in the first sample scheme and the sample schemes in the additional iterations are selected based on areas of relatively high variance and areas where the predicted resource outputs of the preceding schemes are relatively high.
Optionally, the multiple parameters for SAGD resource production further include one or more of number of the well pairs, relative distances between the well pairs, locations of the well pairs, steam injection rates into the well pairs, steam quality, steam temperature, degree of sub-cool in the well pairs, porosity of a subterranean reservoir, and permeability of the subterranean reservoir.
Optionally, each well pair includes an injection well and a production well extending relatively horizontal and parallel to each other along a length through a subterranean reservoir. The injection well is disposed vertically between the production well and a ground surface above the subterranean reservoir. The injection well is configured to receive steam and inject the steam into the subterranean reservoir. The production well is configured to collect an oil-based resource from the subterranean reservoir that is heated by the steam.
Optionally, the stop criterion is based on at least one of a number of iterations, expiration of a designated time period, or a measured variance in the predicted resource outputs associated with the sample schemes in a group of recent iterations being less than a designated threshold value.
Optionally, the stop criterion is met responsive to the measured variance in the predicted resource outputs associated with the sample schemes in the group of recent iterations being less than the designated threshold value. The group includes the sample schemes from a previous three iterations before the stop criterion is met. The designated threshold value is no greater than 5%.
In another embodiment, a system is provided that includes a controller configured to evaluate different parameters for allocating steam to well pairs in a well pad for steam assisted gravity drainage (SAGD) resource production. The controller is configured to determine expected resource outputs from the well pad based on one or more steam introduction parameters. The steam introduction parameters include one or more of different quantities of steam directed into the well pairs, different pressures controlled within the well pairs, or different time periods that the steam is directed into the well pairs. The controller is also configured to control introduction of the steam into the well pairs using at least one of the steam introduction parameters based on the expected resource outputs.
It is to be understood that the above description is intended to be illustrative, and not restrictive. For example, the above-described embodiments (and/or aspects thereof) may be used in combination with each other. In addition, many modifications may be made to adapt a particular situation or material to the teachings of the inventive subject matter without departing from its scope. While the dimensions and types of materials described herein are intended to define the parameters of the inventive subject matter, they are by no means limiting and are example embodiments. Many other embodiments will be apparent to those of ordinary skill in the art upon reviewing the above description. The scope of the inventive subject matter should, therefore, be determined with reference to the appended claims, along with the full scope of equivalents to which such claims are entitled. In the appended claims, the terms “including” and “in which” are used as the plain-English equivalents of the respective terms “comprising” and “wherein.” Moreover, in the following claims, the terms “first,” “second,” and “third,” etc. are used merely as labels, and are not intended to impose numerical requirements on their objects. Further, the limitations of the following claims are not written in means-plus-function format and are not intended to be interpreted based on 35 U.S.C. § 112(f), unless and until such claim limitations expressly use the phrase “means for” followed by a statement of function void of further structure.
This written description uses examples to disclose several embodiments of the inventive subject matter and also to enable any person of ordinary skill in the art to practice the embodiments of the inventive subject matter, including making and using any devices or systems and performing any incorporated methods. The patentable scope of the inventive subject matter is defined by the claims, and may include other examples that occur to those of ordinary skill in the art. Such other examples are intended to be within the scope of the claims if they have structural elements that do not differ from the literal language of the claims, or if they include equivalent structural elements with insubstantial differences from the literal languages of the claims.
The foregoing description of certain embodiments of the inventive subject matter will be better understood when read in conjunction with the appended drawings. To the extent that the figures illustrate diagrams of the functional blocks of various embodiments, the functional blocks are not necessarily indicative of the division between hardware circuitry. Thus, for example, one or more of the functional blocks (for example, processors or memories) may be implemented in a single piece of hardware (for example, a general purpose signal processor, microcontroller, random access memory, hard disk, and the like). Similarly, the programs may be stand-alone programs, may be incorporated as subroutines in an operating system, may be functions in an installed software package, and the like. The various embodiments are not limited to the arrangements and instrumentality shown in the drawings.
As used herein, an element or step recited in the singular and proceeded with the word “a” or “an” should be understood as not excluding plural of said elements or steps, unless such exclusion is explicitly stated. Furthermore, references to “one embodiment” of the inventive subject matter are not intended to be interpreted as excluding the existence of additional embodiments that also incorporate the recited features. Moreover, unless explicitly stated to the contrary, embodiments “comprising,” “including,” or “having” an element or a plurality of elements having a particular property may include additional such elements not having that property.
Since certain changes may be made in the above-described systems and methods for communicating data in a vehicle consist, without departing from the spirit and scope of the inventive subject matter herein involved, it is intended that all of the subject matter of the above description or shown in the accompanying drawings shall be interpreted merely as examples illustrating the inventive concept herein and shall not be construed as limiting the inventive subject matter.
| Filing Document | Filing Date | Country | Kind |
|---|---|---|---|
| PCT/US16/49645 | 8/31/2016 | WO | 00 |
