Pump control frame
A local edge framework with machine learning models dynamically adjusts pump operations to address changing conditions, enhancing efficiency and extending the life of oil and gas production pumps by identifying and mitigating risks in real-time.
Patent Information
- Application Number
- CN202380036114.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Priority Date
- 2022-02-25
- Filing Date
- 2023-02-23
- Publication Date
- 2024-12-06
AI Technical Summary
Existing pump systems in oil and gas production face challenges in efficiently managing dynamic changes in surface and subsurface conditions, leading to potential catastrophic failures due to static operating thresholds that do not account for gradual changes, such as increased solids production or gas intake, resulting in reduced operational efficiency and increased maintenance costs.
Implementing a local edge framework with machine learning models to analyze pump data and dynamically adjust operating conditions, using data-driven techniques to identify and mitigate risks, thereby optimizing pump performance and extending its useful life.
The edge framework enhances pump operation by detecting early signs of potential failures, improving efficiency, reducing downtime, and extending the operational life of pumps by adapting to real-time changes in conditions, thus optimizing production and reducing maintenance costs.
Smart Images

Figure CN119096167A_ABST
Abstract
Description
[0001] CROSS-REFERENCE TO RELATED APPLICATIONS
[0002] The present disclosure claims priority to U.S. Provisional Application No. 63 / 313,951, filed on February 25, 2022, which is incorporated herein by reference in its entirety. Background Art
[0003] Various techniques can be used for artificial lift, for example, to facilitate the production of fluids from a reservoir. In various cases, artificial lift may be intended to increase the production of liquids from a reservoir; in other cases, it may increase the production of gas. Various artificial lift techniques may employ a pump driven by an electric motor, which may be located at the surface or downhole. Summary of the invention
[0004] A method may include receiving data for a downhole pump operation utilizing a device including a pump; analyzing the data utilizing a local edge framework, the local edge framework executing a trained machine learning model to identify an operating condition of the downhole pump operation associated with a risk of a shortened useful life of the pump; and issuing instructions to the device to resolve the operating condition. A system may include a processor; a memory accessible to the processor; processor-executable instructions stored in the memory and executable by the processor to instruct the system to: receive data for a downhole pump operation utilizing a device including a pump; analyzing the data utilizing a local edge framework, the local edge framework executing a trained machine learning model to identify an operating condition of the downhole pump operation associated with a risk of a shortened useful life of the pump; and issuing instructions to the device to resolve the operating condition. One or more computer-readable storage media may include processor-executable instructions to instruct a computing system to: receive data for a downhole pump operation utilizing a device including a pump; analyzing the data utilizing a local edge framework, the local edge framework executing a trained machine learning model to identify an operating condition of the downhole pump operation associated with a risk of a shortened useful life of the pump; and issuing instructions to the device to resolve the operating condition. Various other devices, systems, methods, etc. are also disclosed.
[0005] This Summary is provided to introduce a selection of concepts that are further described below in the Detailed Description. This Summary is not intended to identify key or essential features of the claimed subject matter, nor is it intended to be used as an aid in limiting the scope of the claimed subject matter. BRIEF DESCRIPTION OF THE DRAWINGS
[0006] The features and advantages of the embodiments may be more readily understood by referring to the following description in conjunction with the accompanying drawings.
[0007] Figure 1 An example of a system is shown;
[0008] Figure 2 An example of a system is shown;
[0009] Figure 3 An example of a system is shown;
[0010] Figure 4 An example of a system is shown;
[0011] Figure 5 An example of the method is shown;
[0012] Figure 6 Examples of frameworks and exemplary techniques are shown;
[0013] Figure 7 A series of exemplary plots are shown;
[0014] Figure 8 A series of exemplary plots are shown;
[0015] Fig. 9 An example of a drawing is shown;
[0016] Fig.10 An exemplary drawing is shown;
[0017] Fig.11 A series of exemplary plots are shown;
[0018] Fig.12 A series of exemplary plots are shown;
[0019] Fig.13 An example of the method is shown;
[0020] Fig.14 An example of a method is shown;
[0021] Fig.15 An example of a drawing is shown;
[0022] Fig.16 An example of a scene and an example of a controller are shown;
[0023] Fig.17 An example of the method is shown;
[0024] Fig.18 Examples of computers and network equipment are shown; and
[0025] Fig.19 Exemplary components of systems and networked systems are shown. DETAILED DESCRIPTION
[0026] The following description includes the best mode currently contemplated for practicing the described embodiments. This description should not be understood in a limiting sense, but is merely intended to describe the general principles of the embodiments. Reference should be made to the published claims to determine the scope of the described embodiments.
[0027] Figure 1 An example of a system 100 is shown that includes a workspace framework 110 that can be used for instantiation, presentation, interaction with, etc. of a graphical user interface (GUI) 120. Figure 1 In the example of , GUI 120 may include graphical controls for computing framework (eg, application) 121 , project 122 , visualization 123 , one or more other features 124 , data access 125 , and data storage 126 .
[0028] exist Figure 1 In the example of, the workspace framework 110 can be customized for a specific geological environment (such as the exemplary geological environment 150). For example, the geological environment 150 may include multiple layers (e.g., stratification) that include a reservoir 151 and may intersect a fault 153. As an example, the geological environment 150 may be equipped with various sensors, detectors, actuators, and the like. For example, the device 152 may include communication circuits for receiving and transmitting information relative to one or more networks 155. Such information may include information associated with downhole equipment 154, which may be equipment for collecting information, assisting in resource recovery, and the like. Other equipment 156 may be located remotely from the well site and include sensing, detection, transmission, or other circuits. Such equipment may include storage and communication circuits to store and transmit data, instructions, and the like. As an example, one or more satellites may be provided for purposes of communication, data collection, and the like. For example, Figure 1 The satellite is shown in communication with a network 155 which may be configured for communications, and it should be noted that the satellite may additionally or alternatively include circuitry for imaging (eg, spatial, spectral, temporal, radiometric, etc.).
[0029] Figure 1The geological environment 150 is also shown as optionally including equipment 157 and 158 associated with a well, the well including a substantially horizontal portion that may intersect one or more fractures 159. For example, consider a well in a shale formation, the shale formation may include natural fractures, artificial fractures (e.g., hydraulic fractures), or a combination of natural fractures and artificial fractures. As an example, a laterally extending reservoir may be drilled. In such an example, there may be lateral changes in properties, stresses, etc., where an assessment of such changes may help plan, operate, etc. to develop a laterally extending reservoir (e.g., via fracturing, injection, extraction, etc.). As an example, equipment 157 and / or 158 may include components, one or more systems, etc. for fracturing, seismic sensing, seismic data analysis, assessment of one or more fractures, etc.
[0030] exist Figure 1 In the example of FIG. 1 , the GUI 120 shows various features of a computing environment, which may include various features of a DELFI environment, which may be referred to as a DELFI framework, which may be one of a plurality of frameworks. As an example, the DELFI framework may include various other frameworks, which may include, for example, one or more types of models (e.g., simulation models, etc.). Some examples of frameworks may include the DRILLPLAN, PETREL, TECHLOG, PIPESIM, ECLIPSE, and INTERSECT frameworks (Schlumberger, Inc., Houston, Texas).
[0031] The DRILLPLAN framework is used for digital well construction planning and includes features for automation of repetitive tasks and validation workflows, thereby enabling rapid generation of drilling plans (eg, digital drill plans, etc.) of improved quality while ensuring consistency.
[0032] The PETREL framework is part of the DELFI cognitive E&P environment (Schlumberger Ltd., Houston, Texas) and is used in geoscience and geoengineering, for example, to analyze subsurface data from exploration, development, drilling to reservoir fluid production.
[0033] The TECHLOG framework can handle and process field and laboratory data from various geological environments (e.g., deepwater exploration, shale, etc.). The TECHLOG framework can structure wellbore data for analysis, planning, etc.
[0034] The PIPESIM simulator includes a solver that can provide simulation results, such as, for example, multiphase flow results (e.g., from the reservoir to the wellhead and beyond, etc.), flow lines and surface facility performance, etc. The PIPESIM simulator can be integrated, for example, with the AVOCET production operations framework (Schlumberger, Inc., Houston, Texas). As an example, one or more reservoirs can be simulated with respect to one or more enhanced recovery technologies (e.g., considering thermal processes such as steam assisted gravity drainage (SAGD), etc.). As an example, the PIPESIM simulator can be an optimizer that can optimize one or more operating scenarios at least in part via simulation of physical phenomena.
[0035] The ECLIPSE framework provides numerical solutions for reservoir simulators (eg, as a computational framework) for quickly and accurately predicting the dynamic behavior of various types of reservoirs and development scenarios.
[0036] The INTERSECT framework provides a high-resolution reservoir simulator for simulating detailed geological features and quantifying uncertainties, for example, by creating accurate production scenarios, and by integrating accurate models of surface facilities and field operations, the INTERSECT framework can produce reliable results that can be continuously updated through real-time data exchange (e.g., from one or more types of data acquisition equipment in the field that can acquire data during one or more types of field operations, etc.). The INTERSECT framework can provide completion configurations for complex wells, where such configurations can be built in the field; can provide detailed chemical enhanced oil recovery (EOR) recipes, where such recipes can be implemented in the field; can analyze the application of steam injection and other thermal EOR technologies for field implementation, advanced production control in terms of reservoir coupling and flexible field management, and the flexibility to write customized solutions for improved modeling and field management control. As with other example frameworks, the INTERSECT framework can be used as part of the DELFI cognitive E&P environment, for example, for rapid simulation of multiple concurrent cases. For example, a workflow can utilize one or more of the DELFI on-demand reservoir simulation features.
[0037] The aforementioned DELFI environment provides various features for workflows related to subsurface analysis, planning, construction, and production, as shown, for example, in the workspace framework 110. Figure 1 As shown, outputs from the workspace framework 110 can be used to guide, control, etc., one or more processes in the geological environment 150, and feedback 160 (e.g., acquired data regarding operating conditions, equipment conditions, environmental conditions, etc.) can be received via one or more interfaces in one or more forms.
[0038] As an example, the workflow may proceed to a geology and geophysics ("G&G") service provider, which may generate well trajectories, which may involve executing one or more G&G software packages.
[0039] exist Figure 1 In the example of , visualization features 123 can be implemented via workspace framework 110, for example, to perform tasks associated with one or more of a subsurface region, planning operations, constructing wells and / or surface fluid networks, and producing from a reservoir.
[0040] As an example, visualization features may be used to visualize various earth models, properties, and the like in one or more dimensions. As an example, visualization features may be used to present information in multiple dimensions, which presentation may optionally include multi-resolution presentation. In such examples, the information being presented may be associated with one or more frameworks and / or one or more data stores. As an example, visualization features may include one or more control features for controlling equipment, which may include, for example, field equipment that may perform one or more field operations. As an example, a workflow may utilize one or more frameworks to generate information that may be used to control one or more types of field equipment (e.g., drilling equipment, cable equipment, fracturing equipment, etc.).
[0041] With respect to reservoir models that may be suitable for use by the simulator, consider seismic data acquisition, such as acquired via reflection seismology, which can be used in geophysics, for example, to estimate the properties of underground strata. As an example, reflection seismology can provide seismic data that represents elastic energy waves (e.g., as transmitted by P-waves and S-waves, in a frequency range of about 1 Hz to about 100 Hz). Seismic data can be processed and interpreted, for example, to better understand the composition, fluid content, range, and geometry of underground rocks. Such interpretation results can be used to plan, simulate, perform, etc., one or more operations for extracting fluids from a reservoir (e.g., reservoir rock, etc.).
[0042] Field acquisition equipment can be used to acquire seismic data, which can be in the form of a track, where the track can include values organized with respect to time and / or depth (e.g., considering 1D, 2D, 3D, or 4D seismic data). For example, consider an acquisition device that acquires digital samples at a rate of one sample every about 4 ms. Given the speed of sound in one or more media, the sampling rate can be converted to an approximate distance. For example, the speed of sound in rock can be about 5 km per second. Therefore, a sampling time interval of about 4 ms will correspond to a sampling "depth" interval of about 10 meters (e.g., assuming a path length from the source to the boundary and from the boundary to the sensor). As an example, the duration of the track may be about 4 seconds; therefore, for a sampling rate of one sample at about 4 ms intervals, such a track will include about 1000 samples, where the samples acquired later correspond to deeper reflective boundaries. If the 4-second track duration of the above example is divided by two (e.g., to take into account reflections), the deepest boundary depth can be estimated to be about 10 km for vertically aligned sources and sensors (e.g., assuming a speed of sound of about 5 km per second).
[0043] As an example, the simulator can utilize various types of constructions, which can be referred to as entities. Entities can include earth entities or geological objects, such as wells, ground, reservoirs, etc. Entities can include virtual representations of actual physical entities that may be reconstructed for simulation purposes. Entities can include entities based on data obtained via sensing, observation, etc. (e.g., considering entities based at least in part on seismic data and / or other information). As an example, an entity can be characterized by one or more properties (e.g., a geometric pillar grid entity of an earth model can be characterized by porosity properties, etc.). Such properties can represent one or more measurements (e.g., acquired data), calculations, etc.
[0044] As an example, a simulator can utilize an object-based software framework that can include entities based on predefined classes to facilitate modeling and simulation. As an example, an object class can encapsulate reusable code and associated data structures. Object classes can be used to instantiate object instances for use by programs, scripts, and the like. For example, a wellbore class can define objects for representing wellbores based on well data. A model of a basin, reservoir, and the like can include one or more wellbores, which can be used, for example, for measurement, injection, production, and the like. As an example, a wellbore can be a wellbore of a well, which can be a completion (e.g., for producing resources from a reservoir, for injecting material, and the like). Although in Figure 1 Several simulators are shown in the example of , but one or more other simulators may additionally or alternatively be used.
[0045] As mentioned, the framework can be implemented within or in an operably connected manner to the DELFI cognitive exploration and production (E&P) environment (SLB in Houston, Texas), which is a secure, cognitive, cloud-based collaborative environment that integrates data and workflows using digital technologies (such as artificial intelligence and machine learning). As an example, such an environment can provide for operations involving one or more frameworks. The DELFI environment can be referred to as the DELFI framework, which can be one of a plurality of frameworks. As an example, the DELFI framework can include various other frameworks, which can include, for example, one or more types of models (e.g., simulation models, etc.).
[0046] Figure 2 An example of a system 200 in a geological environment is shown. As shown, the system 200 can include a progressive cavity pump (PCP) 210, which includes a stator 214 and a rotor 218, wherein the stator 214 includes a liner (e.g., an elastomer, etc.) that forms a cavity relative to the rotor 218. In such an example, when the rotor 218 is rotated by a drive head 240, the rotor 218 can be used to pump fluid, for example, pump fluid upward in the cavity. In some cases, a so-called left-handed helical PCP for flowing fluid downward can be used in addition to or instead of a right-handed helical PCP for flowing fluid upward.
[0047] exist Figure 2 In an example of the embodiment of the present invention, the drive head 240 is coupled to a transmission 242, which is coupled to an electric motor 244. In such an example, the transmission 242 may include one or more belts, one or more gears, etc. The electric motor 244 may be electrically coupled to a power source 250 and, for example, a controller 260. In such an example, the controller 260 may control the supply of power from the power source 250 to the electric motor 244 to control the rotation of the rotor 218 of the PCP 210.
[0048] exist Figure 2 In an example of a well, system 200 may include surface equipment, such as, for example, separation equipment 270, which may include one or more solid chambers 280. In such an example, separation equipment 270 may be used to separate solids produced from the well and collect them in one or more solid chambers 280. In such an example, the solid chamber may include a sensor that may provide information about the level of solids collected in the solid chamber. For example, once the solid chamber is full, it may be emptied, which may be a manual process requiring human intervention.
[0049] exist Figure 2In the example, the geological environment is shown as a coal bed methane geological environment. Coal bed methane is natural gas that is present in coal mines, generally about 300 to 600 meters underground. During the formation of coal, large amounts of gas are produced and stored on the internal surface of the coal. Since coal has a relatively large internal surface area, it can store seven times the amount of natural gas as a conventional natural gas reservoir of the same volume of rock.
[0050] Coal bed gas is held in place by water pressure. To extract the coal bed gas, wells can be drilled through the coal seams and the water pressure can be reduced by extracting some of the water. Figure 2 As shown in the example of , the reduction in water pressure helps to release natural gas from the coal. As an example, the natural gas and water can be separated so that the natural gas can be piped to a compression plant for transportation via a natural gas transmission pipeline. As an example, a gas turbine generator can be used as a power source, where a portion of the gas is burned to drive a gas turbine coupled to the generator to generate electricity. For example, the power source 250 can be a natural gas turbine generator. As an example, the controller 260 can be operably coupled to the natural gas turbine generator to control the generation of electricity to drive the electric motor 244 and thereby drive the PCP 210. As an example, a single power source can be operably coupled to one or more electric motors to drive one or more PCPs, etc.
[0051] In some cases, hydraulic fracturing can be used to facilitate the extraction of coalbed methane. For example, consider the injection of high-pressure fluids into a coal seam to enlarge existing fractures and create new fractures. Proppants such as sand can be mixed with the injected fluid, carried into the fractures, and kept open after the fracture treatment is completed and the pressure is released. This process can enhance water removal and coalbed methane extraction.
[0052] As an example, in an operation such as Figure 2 A KUDU PCP (Schlumberger Inc., Houston, Texas) was used in the operation shown in FIG. 1 . As an example, a PCP having a particular size, geometry, volume fraction, etc. may be selected.
[0053] As an example, the production process may optionally utilize one or more fluid pumps, such as, for example, a PCP, an electric submersible pump (e.g., consider a centrifugal pump, a rod pump, etc.). As an example, the production process may implement one or more so-called "artificial lift" (or artificial-lift) techniques. Artificial lift techniques may operate by adding energy to a fluid, for example, to initiate, enhance, etc., production of the fluid.
[0054] Figure 3An example of an ESP system 300 is shown, which includes an ESP 310 as an example of a device that can be placed in a geological environment. As an example, the ESP can be expected to function in a certain environment for a long period of time (e.g., optionally, a period of several years). As an example, one or more commercially available ESPs (such as the REDA ESP of Schlumberger, Inc. of Houston, Texas) can be used for various operations.
[0055] exist Figure 3 In the example of FIG. 3 , ESP system 300 includes network 301, well 303 (e.g., with surface equipment, etc.) set in a geological environment, power supply 305, ESP 310, controller 330, motor controller 350, and VSD unit 370. Power supply 305 can receive power from a power grid, an on-site generator (e.g., a turbine driven by natural gas), or other sources. Power supply 205 can provide, for example, a voltage of about 4.16 kV.
[0056] As shown, well 303 includes a wellhead, which may include a choke (e.g., a throttle valve). For example, well 303 may include a throttle valve to control various operations, such as reducing the pressure of the fluid from the high pressure in the closed wellbore to atmospheric pressure. The adjustable flow valve includes a valve that is constructed to resist the wear caused by high-speed, solid-containing fluid flow by limiting or sealing elements. The wellhead may include one or more sensors, such as a temperature sensor, a pressure sensor, a solid sensor, etc.
[0057] With respect to ESP 310, it is shown as including multiple cables 311 (e.g., or one cable), pump 312, gas processing features 313, pump inlet 314, motor 315, one or more instrument / sensor units 316 (e.g., temperature, pressure, strain, current leakage, vibration, etc.), and optional protectors 317. With respect to an example of an instrument / sensor unit, consider the PHOENIX meter (Schlumberger Inc., Houston, Texas).
[0058] As an example, an ESP motor may include a three-phase squirrel cage with two-pole induction. As an example, an ESP motor may include steel stator laminations that may help focus magnetic forces on the rotor, for example, thereby helping to reduce energy losses. For example, the stator winding may include copper and an insulating layer.
[0059] exist Figure 3 In the example of , well 303 may include one or more well sensors 320. Figure 3In the example, the controller 330 may include one or more interfaces, such as for receiving, transmitting, or receiving and transmitting information with a motor controller 350, a VSD unit 370, a power source 305 (e.g., a gas turbine generator, a power company, etc.), a network 301, equipment 303 in the well, equipment in another well, etc.
[0060] like Figure 3 As shown, controller 330 may include or provide access to one or more frameworks. In addition, controller 330 may include features of an ESP motor controller and optionally replace ESP motor controller 350. Figure 3 In an example of the motor controller 350, the motor controller 350 can be a commercially available motor controller, such as a UNICONN motor controller (Schlumberger, Inc., Houston, Texas), which can be connected to a SCADA system. For example, the motor controller 350 can perform some control and data acquisition tasks for an ESP, surface pump, or other monitoring well. For example, the UNICONN motor controller can interface with a PHOENIX monitoring system, for example, to obtain pressure, temperature, and vibration data and various protection parameters, and to provide DC power to downhole sensors (e.g., sensors of instruments, etc.). The UNICONN motor controller can interface with a fixed speed drive (FSD) controller or a VSD unit (such as VSD unit 370).
[0061] For VSD units, UNICONN motor controllers can monitor VSD output current, ESP operating current, VSD output voltage, supply voltage, VSD input and VSD output power, VSD output frequency, drive load, motor load, three-phase ESP operating current, three-phase VSD input or output voltage, ESP rotation frequency and phase-to-ground voltage.
[0062] exist Figure 3 In the example of , the ESP motor controller 350 includes various modules to handle, for example, reverse rotation of the ESP, sand plugging of the ESP, flux of the ESP, and gas lock of the ESP. The motor controller 350 may additionally, alternatively, include any of a variety of features, etc.
[0063] Regarding motor 315, consider, for example, a REDA MAXIMUS PRO MOTOR electric motor (Schlumberger, Inc., Houston, Texas), which can be a 387 / 456 series with a housing outer diameter of approximately 12 cm, or a 540 / 562 series with a housing outer diameter of approximately 14 cm. As an example, consider a carbon steel housing, a high nickel alloy housing, etc. As an example, consider an operating frequency of approximately 30 to approximately 90 Hz. As an example, consider a maximum winding operating temperature of approximately 200 degrees Celsius. As an example, consider self-lubricating and polymer-lined head and bottom radial bearings. As an example, consider a cable terminal head including a cable connector for electrically connecting a power cable to the motor.
[0064] exist Figure 3 In an example, a shaft segment of pump 312 may be coupled to a shaft segment of protector 317 via a connector, and a shaft segment of protector 317 may be coupled to a shaft segment of motor 315 via a connector. As an example, the ESP may be oriented toward a desired direction, which may be vertical, horizontal, or at another angle (e.g., which may be defined relative to gravity, etc.). The orientation of the ESP relative to gravity may be considered, for example, as a factor for determining ESP characteristics, operation, etc.
[0065] like Figure 3 As shown, motor 315 is an electric motor including a cable connector, for example, for operably coupling the electric motor to a multiphase power cable, for example, optionally via one or more motor lead extensions. The power supplied to motor 315 via the cable connector can also be supplied to instrument / sensor unit 316, for example, via the star point of motor 315 (e.g., the star point of a multiphase motor). This approach can also provide for information transmission from unit 316 to the ground. As an example, unit 316 can include a transmission circuit that can transmit information via the star point of motor 315 and via one or more of cables 311, where such information can be received at a ground unit, etc. (e.g., considering a choke, etc. that can extract information from one or more multiphase electrical conductors, etc.).
[0066] Figure 2 and Figure 3 Systems 200 and 300 are examples of pumping systems that can be deployed on-site at relatively remote locations. For example, consider an offshore location that can be reached by ship or helicopter, or a land location that is far away from densely populated areas so that transportation may be inconvenient, etc.
[0067] As an example, the framework may provide monitoring, control, and optimization of one or more types of pumps (eg, PCP, ESP, etc.), for example, with the goal of increasing operating life and reducing operating costs.
[0068] In various operations, the workflow may include setting operational thresholds based on pump characteristics, which may be primarily based on measured physical characteristics. For example, consider a SCADA-based monitoring system that is used to acquire pump data, where one or more thresholds are set on a programmable logic controller (PLC) to control the pump. In this approach, certain pump optimization algorithms can be run on a server, where these algorithms can be used to further optimize the operational life of the pump. This setup relies on a robust communication network between the pump and the server and the computing power of the server to process such algorithms. However, when there are multiple pumps to monitor, end users are often faced with the challenge of prioritizing the pumps to be addressed (e.g., by problem severity).
[0069] As an example, the framework may be an edge framework for providing field execution for pumps and pump-related activities. For example, consider a PCP edge framework that can monitor the status of a PCP and perform real-time analysis to optimize production and mitigate disruptive situations. As described above, PCPs tend to operate within fixed operating thresholds that, once set, do not change until the pump is replaced. A PCP edge framework (PCP EF) may utilize dynamic thresholds that are determined using data-driven techniques that may be combined with the physics of the operation. In such an example, the PCP EF may operate a PCP or a fleet of PCPs in an intelligent manner by taking into account surface and subsurface factors. This approach may improve the handling of dynamic events, such as increased solids production, high gas intake, low liquid production, etc. Although PCP EFs are mentioned, ESP EFs may be provided that may improve the operation of ESP systems (e.g., see Figure 3 ).
[0070] As an example, the PCP EF can perform 24-hour monitoring of PCP parameters as well as production parameters, calculate correlation coefficients between two or more parameters of interest in real time and render them graphically (e.g., bubble charts, etc.) to easily identify candidate parameters that are not behaving as expected in real time; provide remote control of wells and autonomous actions; reduce the distance to and monitoring effort of the oil well site; promote intelligent review through defined inputs; and improve task management in a manner that leads to proactive well management.
[0071] As explained above, pumps (such as PCPs) are operated by operating the pump within a threshold range, which, once set, typically remain static (unchanged) until a subsequent well intervention. However, such thresholds do not take into account the effects of gradually changing surface and subsurface conditions, which may be caused by, among other factors, higher solids production leading to lower liquid intake, differential pressure due to compressor failure downstream of the pump, higher gas intake, sudden breakouts in the operating envelope, etc. If these issues are not addressed promptly, the pump may fail.
[0072] As an example, an EF (e.g., a pump EF) can be used to detect changes early, particularly one or more changes that could have a catastrophic effect on the pump. This approach can improve operations compared to static threshold approaches, which often fail to prevent various types of catastrophic failures.
[0073] As an example, a data-driven dynamic approach can include executing a workflow that utilizes statistical analysis of historical data to identify operating ranges, where the edge framework can include one or more models (e.g., machine learning models, physics-based models, etc.) based on such operating ranges, where the edge framework can be deployed on an edge framework gateway, for example, to monitor, control, and optimize a pump autonomously and optionally via remote communications.
[0074] Figure 4 An example of system 400 and an example of architecture 401 are shown. As shown, architecture 401 can provide one or more workflows about one or more sites for one or more pumps. As an example, architecture 401 can generate one or more results (e.g., behavioral features, classifications, control actions, etc.) that can be used to operate at one or more sites. In such an example, one or more results can be generated locally and / or remotely (e.g., depending on the number of sites, resources, etc.). As shown, architecture 401 may include one or more classification components, one or more control state components (e.g., for control decision making), etc. Architecture 401 may include one or more physical models, one or more machine learning models, etc. As shown, architecture 401 includes an interface for real-time data, an optional interface for temporary data, etc. The result component may include a result interface, wherein the output result may be a notification, alarm, control trigger, control instruction, etc. that can call one or more pieces of equipment to perform one or more actions.
[0075] As shown, the system 400 may include a power source 402 (e.g., solar energy, a generator, a power grid, etc.), which may provide power to an edge framework gateway 410, which may include one or more computing cores 412 and one or more media interfaces 414, which may, for example, receive a computer-readable medium 440, which may include one or more data structures, such as an image 442, a framework 444, and data 446. In such an example, the image 442 may be an operating system image that enables one or more of the one or more cores 412 to establish an operating system environment suitable for executing one or more applications. For example, the framework 444 may be an application suitable for execution in an operating system established in the edge framework gateway 410. As an example, the framework 444 may be suitable for performing tasks associated with the architecture 401.
[0076] exist Figure 4 In the example of EF gateway 810, edge framework gateway 810 (EF gateway) can include one or more types of interfaces suitable for receiving and / or transmitting information. For example, consider one or more wireless interfaces that can provide local communication at a certain site (such as, local communication with one or more of local devices 432, 434 and 436) and / or remote communication with one or more remote sites 452 and 454. As an example, local devices 432, 434 and 436 can include one or more pumps, one or more sensors, etc.
[0077] As an example, EF gateway 410 may be installed at a location some distance from a city, town, etc. In such an example, EF gateway 410 may be accessible via a satellite communication network.
[0078] A communications satellite is an artificial satellite that relays and amplifies radio telecommunication signals via transponders. A satellite communications network may include one or more communications satellites, for example, providing one or more communications channels. As of 2021, there are approximately 2,000 communications satellites in orbit around the Earth, some of which are geostationary satellites above the equator, so that ground stations can permanently aim their satellite dishes at the satellite rather than tracking it.
[0079] High-frequency radio waves used for telecommunication links travel via line-of-sight, which can be hampered by the curve of the Earth. Communications satellites can relay signals around the curve of the Earth, enabling communications between widely separated geographic locations. Communications satellites can use one or more frequencies (e.g., radio, microwave, etc.), where frequency bands can be regulated and allocated.
[0080] Satellite communications tend to be slower and more expensive than other types of electronic communications due to factors such as distance, equipment, deployment, and maintenance. For well sites without other means of communication, satellite communications may be limited in one or more ways. For example, when the controller is to operate in real-time or near real-time, a cloud-based control approach may introduce excessive latency. Figure 4 As shown in the example of , EF gateway 410 can be deployed where it can operate locally with one or more of devices 432, 434, 436, etc., which can be used for control purposes.
[0081] As needed, communication may occur from time to time between the EF gateway 410 and one or more remote sites 452, 454, etc., which may be carried out via satellite communication with tolerable delay and cost. As an example, the CRM 440 may be a removable drive that can be brought to the site via one or more modes of transportation. For example, consider airdrops, where people are delivered via helicopters, airplanes, or ships, etc.
[0082] Regarding airdrops, it is contemplated to throw an electronic device that can be locally activated after landing or while suspended on a parachute and descending to the ground. Such an electronic device can communicate via a local communication system (such as, for example, a local WiFi, Bluetooth, cellular, or other communication system). In such an example, one or more data structures can be transmitted from the electronic device (e.g., including a CRM) to the EF gateway 410. This approach can provide local control, where one or more people may or may not be present. As an example, an autonomous and / or manually controlled vehicle on site can help locate the electronic device and help download its payload to an EF, such as the EF gateway 410. For example, consider a local drone or land vehicle that can locate an airdropped electronic device and retrieve it, and directly and / or indirectly transmit one or more data structures from the electronic device to the EF. In such an example, a drone or land vehicle can establish communication with the electronic device and / or read data from the electronic device so that the data can be transmitted (e.g., to one or more EFs).
[0083] With respect to drones, drones that include one or more features of one or more of the following types of drones are considered: DJI Matrice 210RTK, DJI Matrice 600PRO, Elistair Orion Tethered Drone, Freefly ALTA 8, GT Aeronautics GT380, Skydio 2, Sensefly eBee X, SkyfrontPerimeter 8, Vantage Robotics Snap, Viper Vantage, and Yuneec H920 Plus Tornado. The DJIMatrice 210RTK has a takeoff weight of 6.2g (including battery and maximum 1.2kg payload), a maximum airspeed of 13-30m / s (30–70mph), and a range of 500m–1km under standard radio / video conditions, but it can be integrated with other systems to obtain a greater range from a base station, with a flight time of 15-30 minutes (e.g., depending on battery and payload selection, etc.). As an example, the gateway can be a mobile gateway that includes one or more features of a drone and / or can be a payload of a drone.
[0084] As an example, the system may include and / or provide access to various resources, which may be part of an environment, such as, for example, the DELFI environment (see, e.g. Figure 1 ).
[0085] As an example, the EF may include a license server, a semi-empirical model component, a framework simulation engine (e.g., a PIPESIM engine, etc.), and a REST API, wherein the REST API may receive one or more API calls, e.g., as one or more model requests, calibration requests, simulation requests, etc. As an example, the EF may respond to the API call with an output, wherein such output may be provided to one or more edge applications, pieces of equipment, etc. (e.g., for individual and / or coordinated control of one or more groups of equipment, etc.).
[0086] Referring again to architecture 401 , as explained above, one or more physics-based models may be deployed to the edge for implementation, such as operating in response to real-time data, in response to historical data, and so on.
[0087] like Figure 4 As shown, EF can be executed within a gateway (such as, for example, an AGORA gateway) (e.g., considering one or more processors, memory, etc., which can be deployed as a "box" that can be locally powered and can communicate locally with other devices via one or more interfaces). As an example, the gateway may include one or more features of an AGORA gateway (e.g., v.202, v.402, etc.) and / or another gateway. For example, consider an INTEL ATOM E3930 or E3950 dual-core with DRAM and eMMC and / or SSD. Such a gateway may include a trusted platform module (TPM), which can provide secure and measurable boot support (e.g., via hashing, etc.). The gateway may include one or more interfaces (e.g., Ethernet, RS485 / 422, RS232, etc.). Regarding power, the gateway may consume less than about 100W (e.g., considering less than 10W or less than 20W). As an example, the gateway may include an operating system (e.g., considering LINUX DEBIAN LTS). As an example, the gateway may include a cellular interface (e.g., 4G LTE with global modem / GPS, etc.). As an example, the gateway may include a WIFI interface (e.g., 802.11a / b / g / n). For example, the gateway may operate using AC 100-240V, 50 / 60Hz, or 24VDC. Regarding size, consider a gateway having a protective box with dimensions of approximately 10 inches x 8 inches x 4 inches.
[0088] As an example, a gateway can be part of a drone. For example, consider a mobile gateway that can take off and land where it can land to operably connect with a device to provide control of such device. In such an example, the device can include a landing pad. For example, a drone can be guided to the landing pad, where it can interact with the device to control the device. As an example, a wellhead can include a landing pad, wherein the wellhead can include one or more sensors (e.g., temperature and pressure), and wherein the mobile gateway can include a feature that generates a fluid flow value using information from the one or more sensors. In such an example, the mobile gateway can issue one or more control instructions (e.g., to a throttle, a pump, etc.).
[0089] As an example, a gateway may include hardware (e.g., circuitry) capable of providing drone operations. As an example, a gateway may be a drone controller and a controller for other devices, where the drone controller may locate the gateway (e.g., via drone flight characteristics, etc.) so that the gateway may control the other devices.
[0090] Figure 5 An example of a method 500 that may be performed using EF, such as, for example, framework 444 that may be executed on EF gateway 410, is shown. As shown, method 500 may include a receiving block 510 for receiving data, a scaling block 520 for scaling data, a statistics block 530 for statistically processing data (e.g., scaled data, etc.), a supervised learning block 540 for performing supervised learning of a machine learning model, an unsupervised learning block 550 for performing unsupervised learning of a machine learning model, and an issuing block 560 for issuing one or more instructions that may invoke one or more types of actions.
[0091] exist Figure 5 In the example of, the scaling block 520 can perform scaling on the data relative to other data and / or one or more factors. For example, consider data scaling intended to account for seasonal factors so that seasonal variations can be handled in pump and pump-related data. With respect to the statistics block 530, consider performing cross-plotting, gradient analysis, and / or correlation. With respect to the supervised learning block 540, it can include performing regression analysis to train a machine learning model (ML model). With respect to the unsupervised learning block 550, consider performing cluster analysis by implementing one or more clustering techniques. With respect to the issue block 560, consider a method that can issue one or more types of instructions regarding one or more thresholds, one or more sampling rates, one or more data requirements, and the like.
[0092] As an example, method 500 may provide streaming analysis at the EF, wherein real-time calculations of multiple statistical parameters are performed, such as torque gradients and correlation coefficients of various parameters with respect to rod speed (e.g., pump rotor speed). In such an example, the operating condition of the pump may be monitored at least in part by cross-checking the calculated outputs with one or more set dynamic thresholds.
[0093] Figure 6 An example of a framework 600 is shown, wherein components may be included in a system such as Figure 4 444 of the framework. For example, consider the downstream network component 610, the gas intake component 620, the operating envelope component 630, and the solid production component 640. In such an example, the operating envelope component 630 can provide utilization of torque gradients and / or correlation coefficients that are closely related to pump operation and / or pump failure (e.g., remaining useful life, risk of failure, etc.). As an example, the solid production component 640 can provide determinations regarding remote operation, automatic flushing, etc. With respect to the downstream network component 610, it can provide data regarding casing pressure and gas volume (e.g., information closely related to one or more downstream fluid networks). With respect to the gas intake component 620, it can provide a correlation coefficient relative to downhole gas pressure (DHGP), for example, for an indication of gas intake; it should be noted that gas intake can reduce liquid pumping (e.g., whether for PCP, ESP, etc.).
[0094] As an example, framework 600 may include various components that may be invoked when a threshold is breached, which may include one or more of increasing and / or decreasing pump speed, running a solids control remote and / or automatic flushing process, etc., among other actions.
[0095] As mentioned above, various pumps are operated using static thresholds that do not take into account the dynamic conditions to which the pumps are subjected. Furthermore, such static operations often rely on the uptime of a robust communication network and a high-performance computing setup on the server side. In such cases, automated modes of operation are rarely used, as human intervention is required to resolve the problem.
[0096] As explained above, a method can be implemented using an edge-based application that can utilize a data-driven approach in conjunction with a control mechanism deployed at the edge. This approach can monitor the pump in near real-time conditions and can mitigate one or more types of problems associated with communication network delays or power outages.
[0097] As explained above, an edge-based approach can provide users with the flexibility to resolve issues remotely or autonomously. The dynamic nature of edge-implemented applications helps ensure that pumps operate more optimally, which can extend their operational life. Additionally, since edge-implemented applications can implement control of the pump through local and / or remote actions, the number of trips to the well site can be reduced.
[0098] Figure 7 A series of exemplary data graphs 700 are shown that span a period of several months for PCP operation. Specifically, the data graphs 700 include pump speed (PS) and torque correlation graphs, PS and water flow (FR) correlation graphs, PS and gas flow (FR) correlation graphs, casing pressure and gas volume correlation graphs, and scaled data and graphs of calculated efficiency (e.g., pump efficiency). In the data graphs 700, pump speed is selected for correlation because pump speed can be a controllable variable for pump operation. As explained above, the pump can include a speed-controllable electric motor, wherein the rotor of the electric motor can be directly or indirectly coupled to the rotor of the pump. For example, the electric motor can include a rotor / stator, and the pump can include a rotor / stator (see, e.g., Figure 1 ), or may include a rotor with impeller blades configured with a diffuser (see e.g. Figure 2 ); It should be noted that other types of pumps may utilize other pump configurations (e.g., a linear rod pump that can reciprocate at a controlled speed, etc.).
[0099] The various data in the data graph 700 are processed using regression techniques to generate regression information and cross-correlation techniques to generate correlation coefficients, as set forth in Table 1 below.
[0100] Table 1 Correlation coefficients
[0101] return Pump speed Gas volume Water FR Casing pressure Torque 0.583 0.7626 0.3084 0.1152 Scrolling relevance PS-Torque PS-Gas FR PS-Water FR PS-Casing (Time = 4 days) 0.553 0.567 0.699 0.268
[0102] As indicated in the figure, a rolling correlation may be calculated, where, for example, a suitable time window (e.g., in hours, days, or weeks) may be selected. In Table 1, the highest correlations within the time window are pump speed and water flow. As an example, one or more of the regression results and the rolling correlation results may be used as an indicator of pump operation (e.g., pump behavior, etc.). For example, consider using one or more thresholds for the regression results and / or correlation results that may provide an indication of pump operation, which may be a trigger condition for issuing control and / or other instructions.
[0103] As indicated in Table 1, regressions may be calculated for various measurements. Such regressions may provide an indication of how much a change in one variable will change one or more other variables. For example, consider torque and pump speed, where a change in pump speed may result in a particular change in torque, which may be related to a change in gas volume, water flow, casing pressure, etc. As an example, the regression results may be used on an online basis for comparison, control, alarm, etc. purposes.
[0104] As about Figure 6 As described, the framework can be used to calculate correlation coefficients, which may be closely related to controlling gas intake, operating envelope, etc. This approach helps to discover conditions that may shorten pump life (e.g., risk of pump failure).
[0105] Figure 8 Various exemplary graphs 800 of torque and pump speed for ten wells are shown, with torques ranging from 0 to 800 and pump speeds ranging from 0 to 350. As shown, each of the ten wells can be compared to one or more other wells so that regions of torque and pump speed can be identified and correlated. For example, wells 1, 4, and 8, wells 2, 5, and 9, and wells 3, 6, and 10 form three groups with respect to pump speed ranges within a set torque range. Figure 8 In the example of , well 7 does not fit into these three groups and can therefore be considered to exhibit different behavior.
[0106] In the diagram of Well 1, clusters are indicated by shading, files, and / or hatching. In this approach, clustering techniques can be implemented to determine an appropriate number of clusters, for example, considering a k-means approach (e.g., k-means nearest neighbor, etc.). In the exemplary diagram of Well 1, five clusters are indicated with trends in pump speed and torque. As an example, clusters can be evaluated for pump operation, where one or more clusters can be used to define suitable operating conditions, such as, for example, an operating envelope (e.g., considering pump speed and torque). In such an example, the operating envelope can be based on a single well or multiple wells. As an example, clustering can be performed using supervised learning and / or unsupervised learning. With respect to unsupervised learning, incoming data can be automatically analyzed, for example via clustering, to determine dynamic operating modes (e.g., envelopes, etc.). In this approach, dynamic control can be implemented in an effort to maintain the operation of the pump within a certain envelope.
[0107] Fig. 9 An exemplary graph 910 of torque gradient (solid fill line) and gas flow gradient (white fill line) versus time is shown, as well as a zoomed-in portion 920 of the torque gradient versus time in the graph 910. Fig. 9In the example of FIG. 920 , torque gradients at various times can be identified, which can be associated with gas flow. For example, the gas flow decreases and then increases, and then stabilizes after multiple changes in the torque gradient value. As indicated in FIG. 920 , various times of torque peaks can be identified, which may occur within approximately 20 hours (e.g., from 23:41 on day 11 to 19:13 on day 12).
[0108] exist Fig. 9 In the example of , graph 920 shows a period of time without gradient data, which corresponds to no operation (e.g., failure, workover time, etc.). As shown, the pump has been tested where torque peaks became frequent, however, limited to some extent. Thereafter, pump operation using the same pump (e.g., or a different pump) returned to proper operation as indicated by the torque gradient.
[0109] As explained above, the torque gradient can be used as an indicator to characterize the operation of the pump. Fig. 9 As shown, such data may be supplemented with respect to one or more other types of data (e.g., gas flow data, etc.). In exemplary diagram 920, the horizontal dashed line may represent a limit (e.g., a threshold) of a torque gradient, which may be based on automatically acquired torque data and processed locally by the edge framework gateway.
[0110] Fig.10 Exemplary data graphs 1010 and 1020 are shown relating to data acquisition (specifically, the frequency or sampling rate of data acquisition). As shown, graph 1010 shows data sampled at 10 second intervals, while graph 1020 shows data sampled at 20 minute intervals. It can be seen that the higher frequency sampling provides an indication of improved torque behavior. As with respect to Figure 6 As illustrated in the framework 600 of FIG. 1 , torque and torque gradient can be related to an operating envelope, where, for example, a correlation coefficient of one or more operating parameters can be calculated for torque and / or torque gradient. The correlation coefficient (e.g., see Table 1) can provide information about trends, behaviors, etc., which can help make operating decisions (e.g., extend the life of a pump, etc.).
[0111] Fig.11 A series of exemplary data graphs 1100 are shown that span a period of several days for PCP operation. Specifically, the data graphs 1100 include PS and torque correlation graphs, PS and water flow (FR) correlation graphs, PS and gas flow (FR) correlation graphs, scaled data graphs, and actual data graphs including calculated efficiencies (e.g., pump efficiencies).
[0112] exist Fig.11In the example of FIG. 1 , certain data over a period of time indicates that the PS-torque correlation increases, the PS-water flow correlation decreases, and the PS-gas production correlation increases while the efficiency decreases. Such data may be an indicator of one or more conditions that may result in a shortened pump life. For example, such data may be an indicator of the presence of elevated solids levels, the effects of gas in the pump cavity, etc. Figure 6 As illustrated in framework 600, a gas intake component 620 can invoke one or more actions to address a gas intake problem (e.g., gas content is above a threshold, which may be dynamic), and a solid production component 640 can invoke one or more actions to address a solid presence problem (e.g., solid content is above a threshold, which may be dynamic).
[0113] exist Fig.11 In the example of FIG1100, the data may be locally transmitted to an edge framework (see, e.g. Figure 4 The edge framework gateway of the framework 444, which can provide execution of one or more edge applications). Fig.11 In an example, one or more correlation techniques can be utilized, such as to calculate a correlation coefficient, which can be within one or more windows (e.g., hours, days, weeks, etc.). As explained above, this approach can reveal trends and / or behaviors that can be addressed by one or more control actions. This approach can help provide an operating envelope for the pump that extends the useful life of the pump. As explained above, this approach implemented at the edge can be autonomous, reducing the need for human intervention.
[0114] Fig.12 A series of exemplary data graphs 1200 are shown, which span a period of several days for PCP operation. Specifically, the data graphs 1200 include PS-torque correlation graphs, PS-water flow (FR) correlation graphs, PS-gas flow (FR) correlation graphs, scaled data graphs, and actual data graphs including calculated efficiency.
[0115] exist Fig.12 In the example of , a period of time is identified where torque increases, water flow decreases, and gas volume remains relatively constant (e.g., relatively constant). During that time, efficiency is decreasing. This data can indicate one or more conditions that may affect the life of the pump. As explained above, such as Fig.11 Figure 1100 and Fig.12 The data in Figure 1200 may be acquired by an edge framework gateway executable on the edge framework, which may detect trends and / or behaviors and invoke one or more control actions designed to extend the life of the pump and / or reduce the need for on-site manual intervention.
[0116] Fig.13An example of a method 1300 is shown including a detection block 1310 for detecting solids accumulation (eg, solids content above a threshold, etc.), a call block 1320 for calling to start a desolidification process, and a call block 1330 for calling to use a surface flow meter.
[0117] like Fig.13 As shown in the example of , the detection block 1310 can be used to evaluate the data for trends and / or behaviors, such as, for example, an increase in torque peaks, a decrease in water flow, and / or an increase in solids accumulation at the ground.
[0118] As about Figure 2 As illustrated in the system 200 of FIG. 1 , a station may include a separation device 270, which may include a solid chamber 280. In such an example, the solid chamber 280 may have a fixed volume, and once the solid chamber is full (e.g., reaches a specific solid volume capacity), manual intervention is required to empty the solid chamber 280. As an example, the solid chamber 280 may include one or more sensors that may provide information about solids accumulation, which the edge frame may utilize, for example, to make one or more decisions regarding pump operation.
[0119] exist Fig.13 In an example, call block 1320 may, for example, call a specific process designed to reduce solids that may be present in a pump (e.g., a pump chamber, multiple pump chambers, etc.). As shown, torque can be controlled during the desolidification process, where one or more of cluster analysis and pump design can be considered. With respect to low flow conditions, completion design can be considered. As an example, various situations can be described using Hi and Lo indicators, where Hi is relatively greater than Lo and Lo is relatively less than Hi. As an example, the conditions may be "Hi-Hi", "Hi-Lo", "Lo-Hi", and "Lo-Lo". As an example, Hi and Lo may be corresponding thresholds, and an operating envelope, etc. may be defined. In such an example, the thresholds may be Hi and Li of torque, Hi and Lo of the correlation of torque with respect to another parameter (e.g., pump speed, etc.), and / or Hi and Lo of another parameter (such as, pump speed (e.g., high torque and low pump speed, etc.)).
[0120] In various situations, an operator may increase the pump speed when dealing with a perceived solids problem, thinking that this will resolve the solids problem. However, slowing the pump speed in increments (e.g., 5% to 10% over 30 minutes, etc.) until a slow speed is reached may cause water to accumulate. Once the water accumulates, it can reduce friction to lubricate the pump, allowing the solids to be pumped at a higher pump speed, thereby pumping out the solids in a manner that reduces the risk of damaging the pump (e.g., elastomers and / or other components).
[0121] exist Fig.13 In the example of , call block 1330 may, for example, call a process that utilizes information that may include obtaining information from a surface flow meter. As shown, the process may include: decreasing the PCP rod speed by a specific percentage (e.g., 10%, etc.) increment within a specific time interval (e.g., 10 minutes, etc.) until a specific behavior transition (e.g., less than a Hi-Lo condition) is indicated; running the PCP for a specific time to allow flow accumulation; increasing the PCP rod speed by a specific percentage (e.g., 10%, etc.) increment within a specific time interval (e.g., 10 minutes) until a specific behavior transition (e.g., less than a Hi-Lo condition) is indicated; and running the PCP for a specific time.
[0122] As an example, the edge framework can invoke control of the pump in a manner such that a solids flushing process is automatically performed, optionally according to a schedule. In such an example, the edge framework can include invoking an unscheduled flushing process if one or more conditions are detected. As an example, the schedule can be automatically adjusted based on the conditions.
[0123] Fig.14 An example of a method 1400 is shown including a detection block 1410 for detecting increased gas ingestion and an invocation block 1420 for invoking utilization of rod speed and downhole gas pressure (DHGP) to address the detected increased gas ingestion.
[0124] like Fig.14 As shown, detection block 1410 may include identification of torque spikes, DHGP drops, reduced efficiency (e.g., increased friction), and / or dry run. Fig.14 As shown, call block 1420 may include calling implementation of a process that includes monitoring a correlation coefficient of the PSDHGP, calculating an optimal DHGP, and incrementally (e.g., 10%, etc.) reducing the PCP pole speed until the optimal DHGP is achieved.
[0125] Fig.15 An exemplary graph 1500 is shown, which includes data on gas flow, rod speed, downhole gauge pressure (DHGP), casing pressure, and water flow. As shown in graph 1500, the edge frame can achieve stable DHGP and rod speed, with a change in casing gas pressure of approximately 10 psi, low gas volume, and low water flow. Fig.15 In the example of, an indication of casing pressure changes can be utilized, for example, regarding the risk of pump failure. For example, a sudden change in casing pressure can increase the risk of pump failure.
[0126] As an example, you can use Figure 6 The downstream network component 610 of the framework 600 is used to evaluate the data in the diagram 1500. For example, casing pressure and gas volume may be downstream factors that are closely related to pump operation.
[0127] As explained above, the framework can be implemented using edge computing resources to acquire data at a desired frequency (e.g., sampling rate, etc.). This approach can provide streaming analysis at the edge site and output various instructions that can help extend the life of the pump and / or reduce the need for local on-site manual intervention. As described above, the edge-based approach can automatically call for the implementation of a flushing process to desolidify the pump, for example, which may help maintain the available space in the solids chamber.
[0128] As an example, the framework can be updateable, e.g., via one or more network connections, local delivery, etc. As an example, the framework can be updated in real time.
[0129] As an example, the framework can reduce bandwidth by processing information locally before transmitting it via a network (e.g., satellite, etc.). In such an example, the framework can decide when and / or what type of information to transmit. As explained above, this approach can take advantage of a relatively high frequency of data collection and analyze that data with respect to trends and / or behaviors. In this approach, trends and / or behaviors can be encoded, where the code is transmitted to a remote location (e.g., off-site). In such an example, if additional information is needed, a remote call to the edge framework can request such additional information, which can be evaluated for decision-making purposes, which can include issuing one or more control instructions from the remote site to the local site. This approach can be layered in an effort to reduce the need to dispatch personnel to the local site for intervention.
[0130] Fig.16 An example of a field 1600 is shown, which includes a controller 1610 and a power supply 1620, which can power pumps at multiple wells, where a control scheme can be implemented for each well. In such an example, the control scheme can be provided by an edge framework that can be executed on an edge framework gateway. This approach can consider the power provided by the power supply 1620, which can be an on-site gas turbine generator that generates electricity by burning a portion of the gas produced in one or more wells. In such an example, a failure of the power supply 1620 may cause the pump at the well to shut down. Therefore, the power supply 1620 can be operated in a manner that extends its life and / or reduces the need for human intervention. For example, the operating range of the power supply 1620 can be controlled so that the total power level is not higher than a level that will significantly shorten the life of the power supply 1620. In such an example, the total power demand of the pumps of the well can be limited and allocated accordingly to achieve the desired optimal operation of the pumps as a team.
[0131] As mentioned, one or more machine learning techniques may be used to enhance process operations, process operating environments, communication frameworks, etc. As explained, various types of information may be generated via operation of a communication framework, where such information may be used to train one or more types of machine learning models to generate one or more trained machine learning models, which may be deployed in one or more frameworks, environments.
[0132] Regarding the type of machine learning model, consider one or more of a support vector machine (SVM) model, a k-nearest neighbor (KNN) model, an ensemble classifier model, a neural network (NN) model, etc. As an example, the machine learning model can be a deep learning model (e.g., a deep Boltzmann machine, a deep belief network, a convolutional neural network, a stacked autoencoder, etc.), an ensemble model (e.g., a random forest, a gradient boosting machine, a bootstrap ensemble, adaptive boosting, stacked generalization, a gradient boosting regression tree, etc.), a neural network model (e.g., a radial basis function network, a perceptron, back propagation, a Hopfield network, etc.), a regularization model (e.g., a ridge regression, a least absolute shrinkage and selection operator, an elastic net, a least angle regression), a rule system model (e.g., cubism, one rule, zero rule, repeated incremental pruning to reduce error), a regression model (e.g., linear regression, ordinary least squares regression, stepwise regression, multivariate adaptive regression spline, local scatter smoothing estimation, logistic regression, etc.), a Bayesian model (e.g., naive Bayes, average dependency estimator, Bayesian belief network, Gaussian naive Bayes, multinomial naive Bayes, Bayesian network), decision tree models (e.g., classification and regression tree, iterated binary tree 3 generations, C4.5, C5.0, chi-square automatic interaction detection, decision tree stump, conditional decision tree, M5), dimensionality reduction models (e.g., principal component analysis, partial least squares regression, Sammon map, multidimensional scaling, projection pursuit, principal component regression, partial least squares discriminant analysis, mixed discriminant analysis, quadratic discriminant analysis, regularized discriminant analysis, flexible discriminant analysis, linear discriminant analysis, etc.), instance models (e.g., k-nearest neighbors, learning vector quantization, self-organizing map, local weighted learning, etc.), clustering models (e.g., k-means, k-median, expectation maximization, hierarchical clustering, etc.), etc.
[0133] As an example, a computing framework with libraries, toolboxes, etc. can be used to build machine models, such as, for example, those of the MATLAB framework (MathWorks, Natick, Massachusetts). The MATLAB framework includes toolboxes that provide supervised and unsupervised machine learning algorithms, including support vector machines (SVMs), boosted and bagged decision trees, k-nearest neighbors (KNNs), k-means, k-medoids, hierarchical clustering, Gaussian mixture models, and hidden Markov models. Another MATLAB framework toolbox is the Deep Learning Toolbox (DLT), which provides a framework for designing and implementing deep neural networks using algorithms, pre-trained models, and applications. DLT provides convolutional neural networks (convolutional networks, CNNs) and long short-term memory (LSTM) networks to classify and regress image, time series, and text data. DLT includes features that use custom training loops, shared weights, and automatic differentiation to build network architectures, such as generative adversarial networks (GANs) and Siamese networks. DLT is used to exchange models with various other frameworks.
[0134] As an example, the TENSORFLOW framework (Google LLC, Mountain View, California) can be implemented, which is an open source software library for data flow programming, including a symbolic math library that can be implemented for machine learning applications that can include neural networks. As an example, the CAFFE framework can be implemented, which is a DL framework developed by the Berkeley AI Research Center (BAIR) (UC Berkeley, California). As another example, consider the SCIKIT platform (e.g., scikit-learn), which utilizes the PYTHON programming language. As an example, a framework such as the APOLLO AI framework (APOLLO.AI GmbH, Germany) can be utilized. As an example, a framework such as the PYTORCH framework (Facebook AI Research Lab (FAIR), Facebook, Menlo Park, California) can be used.
[0135] As an example, a training method may include various actions that may be performed on a dataset to train an ML model. As an example, a dataset may be divided into training data and test data, where the test data may be used for evaluation. A method may include cross-validation of parameters and optimal parameters, which may be used for model training.
[0136] The TENSORFLOW framework can run on multiple CPUs and GPUs with optional CUDA extensions (NVIDIA Corporation, Santa Clara, California) and SYCL extensions (Khronos Group Ltd., Beaverton, Oregon) for general-purpose computing on graphics processing units (GPUs). TENSORFLOW can be used on 64-bit LINUX, MACOS (Apple Inc., Cupertino, California), WINDOWS (Microsoft Corporation, Redmond, Washington), and mobile computing platforms including Android (Google LLC, Mountain View, California) and iOS (Apple Inc.) operating system-based platforms.
[0137] TENSORFLOW computations can be represented as stateful data flow graphs; it should be noted that the name TENSORFLOW derives from the operations that such neural networks perform on multidimensional arrays of data. Such arrays can be called "tensors".
[0138] As an example, the device can utilize TENSORFLOW LITE (TFL) or other types of lightweight frameworks. TFL is a set of tools that supports on-device machine learning, where models can run on mobile, embedded, and IoT devices. TFL is optimized for on-device machine learning by addressing latency (no round-trip to the server), privacy (no personal data leaves the device), connectivity (requires an Internet connection), size (reduced model and binary size), and power consumption (e.g., efficient reasoning and lack of network connection). Multi-platform support, covering ANDROID and iOS devices, embedded LINUX, and microcontrollers. Multiple language support, including JAVA, SWIFT, Objective-C, C++, and PYTHON. High performance, with hardware acceleration and model optimization. Machine learning tasks may include, for example, image classification, object detection, pose estimation, question answering, text classification, etc. on multiple platforms.
[0139] Fig.17 An example of method 1700 is shown, which includes: a receiving block 1710 for receiving data for a downhole pump operation utilizing an apparatus including a pump; an analyzing block 1720 for analyzing the data utilizing a local edge framework, the local edge framework executing a trained machine learning model to identify operating conditions of the downhole pump operation associated with a risk of shortening the useful life of the pump; and an issuing block 1730 for issuing instructions to the apparatus to resolve the operating condition.
[0140] In such an example, the operating condition may relate to solids, where, for example, the instructions include solids flushing process instructions. In such an example, the instructions may be to control a speed of an electric motor operably coupled to the pump.
[0141] As an example, the operating condition may involve gas, where for example the instructions include gas release process instructions. In such an example, the instructions may be to control the speed of an electric motor operably coupled to the pump.
[0142] As an example, a method may include an analysis that characterizes torque. For example, consider performing an analysis to identify torque peaks associated with solids accumulation and / or performing an analysis to identify torque peaks associated with gas ingestion.
[0143] As an example, a trained machine learning model can be trained via unsupervised learning and / or supervised learning.
[0144] As an example, a trained machine learning model may include thresholds, where, for example, the thresholds may include at least one dynamic threshold.
[0145] As an example, the command may be or include a pump speed control command.
[0146] As an example, the data may include gas turbine generator data, where, for example, the gas turbine generator data corresponds to operation of a gas turbine generator that generates electricity to operate at least one pump. In such an example, for coal steam gas operation, a portion of the gas generated at least in part via operation of one or more pumps may be used to power the gas turbine generator. In such an example, when natural gas production declines, operation of the gas turbine generator may become an issue, where instructions may be issued to control operation of the gas turbine generator.
[0147] As an example, a method may include coordinating the issuance of control instructions for multiple downhole pump operations. For example, consider the coordination of pumps relative to one or more gas turbine generators that can operate by burning gas produced by at least one of the multiple downhole pump operations.
[0148] As an example, a system may include a processor; a memory accessible to the processor; processor-executable instructions stored in the memory and executable by the processor to instruct the system to: receive data for a downhole pump operation utilizing a device including a pump; analyze the data utilizing a local edge framework that executes a trained machine learning model to identify operating conditions of the downhole pump operation that are associated with a risk of shortening the useful life of the pump; and issue instructions to the device to resolve the operating conditions.
[0149] As an example, one or more computer-readable storage media may include processor-executable instructions to instruct a computing system to: receive data for a downhole pump operation utilizing an apparatus including a pump; analyze the data utilizing a local edge framework that executes a trained machine learning model to identify operating conditions of the downhole pump operation that are associated with a risk of shortening the useful life of the pump; and issue instructions to the apparatus to resolve the operating conditions.
[0150] As an example, a computer program product may include one or more computer-readable storage media that may include processor-executable instructions to instruct a computing system to perform one or more methods and / or one or more portions of a method.
[0151] In some embodiments, one or more methods may be performed by a computing system. Fig.18 An example of a system 1800 is shown that may include one or more computing systems 1801-1, 1801-2, 1801-3, and 1801-4, which may be operatively coupled via one or more networks 1809, which may include wired and / or wireless networks.
[0152] By way of example, a system may include a single computer system or an arrangement of distributed computer systems. Fig.18 In the example of , computer system 1801-1 may include one or more modules 1802, which may be or may include, for example, processor-executable instructions that may be executed to perform various tasks (e.g., receiving information, requesting information, processing information, simulating, outputting information, etc.).
[0153] As an example, the modules may execute independently or in conjunction with one or more processors 1804 that are operably coupled (e.g., via wire, wirelessly, etc.) to one or more storage media 1806. As an example, one or more of the one or more processors 1804 may be operably coupled to at least one of the one or more network interfaces 1807. In such an example, the computer system 1801-1 may transmit and / or receive information, for example, via one or more networks 1809 (e.g., considering one or more of the Internet, a private network, a cellular network, a satellite network, etc.).
[0154] As an example, computer system 1801-1 may receive information from and / or transmit information to one or more other devices, which may be or include, for example, one or more of computer systems 1801-2, etc. The device may be located at a different physical location than computer system 1801-1. As an example, the location may be, for example, a processing facility location, a data center location (e.g., a server farm, etc.), a drilling rig location, a well site location, a downhole location, etc.
[0155] As an example, a processor may be or include a microprocessor, a microcontroller, a processor module or subsystem, a programmable integrated circuit, a programmable gate array, or another control or computing device.
[0156] As an example, storage media 1806 can be implemented as one or more computer-readable or machine-readable storage media. As an example, storage can be distributed within and / or between multiple internal and / or external enclosures of the computing system and / or additional computing systems.
[0157] As examples, the one or more storage media may include one or more different forms of memory, including: semiconductor memory devices, such as dynamic or static random access memory (DRAM or SRAM), erasable and programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), and flash memory; disks, such as fixed disks, floppy disks, and removable disks; other magnetic media, including magnetic tape; optical media, such as compact disks (CDs) or digital video disks (DVDs), Blu-ray disks, or other types of optical storage devices; or other types of storage devices.
[0158] As an example, one or more storage media may be located in the machine running the machine-readable instructions, or located at a remote site from which the machine-readable instructions may be downloaded over a network for execution.
[0159] As examples, various components of a system (such as, for example, a computer system) may be implemented in hardware, software, or a combination of both hardware and software (eg, including firmware), including one or more signal processing circuits and / or application specific integrated circuits.
[0160] As an example, a system may include a processing device, which may be or include a general purpose processor or a dedicated chip (eg, or chipset) such as an ASIC, FPGA, PLD, or other suitable device.
[0161] Fig.19Components of an example of a computing system 1900 and an example of a networked system 1910 with a network 1920 are shown. System 1900 includes one or more processors 1902, a memory and / or storage component 1904, one or more input and / or output devices 1906, and a bus 1908. In an exemplary embodiment, instructions may be stored in one or more computer-readable media (e.g., memory / storage component 1904). Such instructions may be read by one or more processors (e.g., one or more processors 1902) via a communication bus (e.g., bus 1908), which may be wired or wireless. The one or more processors may execute such instructions to implement (in whole or in part) one or more properties (e.g., as part of a method). A user may view output from a process and interact with the process via an I / O device (e.g., device 1906). In an exemplary embodiment, a computer-readable medium may be a storage component, such as a physical memory storage device, such as a chip, a chip on a package, a memory card, etc. (e.g., a computer-readable storage medium).
[0162] In an exemplary embodiment, the components may be distributed, such as distributed in a network system 1910. The network system 1910 includes components 1922-1, 1922-2, 1922-3, ... 1922-N. For example, component 1922-1 may include one or more processors 1902, and one or more components 1922-3 may include a memory accessible by one or more processors 1902. In addition, one or more components 1922-2 may include an I / O device for display and optionally interacting with the method. The network 1920 may be or include the Internet, an intranet, a cellular network, a satellite network, etc.
[0163] As an example, the device can be a mobile device including one or more network interfaces for transmitting information. For example, the mobile device can include a wireless network interface (e.g., operable via IEEE 802.11, ETSI GSM, BLUETOOTH, satellite, etc.). As an example, the mobile device can include components such as a main processor, a memory, a display, a display graphics circuit (e.g., optionally including touch and gesture circuits), a SIM slot, an audio / video circuit, a motion processing circuit (e.g., an accelerometer, a gyroscope), a wireless LAN circuit, a smart card circuit, a transmitter circuit, a GPS circuit, and a battery. As an example, the mobile device can be configured as a cellular phone, a tablet computer, etc. As an example, the method can be implemented (e.g., in whole or in part) using a mobile device. As an example, the system can include one or more mobile devices.
[0164] As an example, the system may be a distributed environment, such as a so-called "cloud" environment, in which various devices, components, etc. interact for purposes of data storage, communication, computing, etc. As an example, a device or system may include one or more components for transmitting information via one or more of the Internet (e.g., in the case of communication via one or more Internet protocols), a cellular network, a satellite network, etc. As an example, the method may be implemented in a distributed environment (e.g., in whole or in part as a cloud-based service).
[0165] As an example, information can be input from a display (e.g., consider a touch screen), output to a display, or both. As an example, information can be output to a projector, a laser device, a printer, etc. so that the information can be viewed. As an example, information can be output stereoscopically or holographically. Regarding printers, consider 2D or 3D printers. As an example, a 3D printer can include one or more substances that can be output to build a 3D object. For example, data can be provided to a 3D printer to build a 3D representation of an underground formation. As an example, layers (e.g., horizons, etc.) can be built in 3D, geological bodies, etc. can be built in 3D. As an example, wellbores, fractures, etc. can be built in 3D (e.g., as positive structures, as negative structures, etc.).
[0166] Although only a few exemplary embodiments have been described in detail above, those skilled in the art will readily appreciate that many modifications may be made in the exemplary embodiments. It is therefore intended that all such modifications be included within the scope of the present disclosure as defined in the appended claims. In the claims, means-plus-function clauses are intended to cover structures described herein as performing the recited function, and not only structural equivalents, but also equivalent structures. Thus, while a nail and a screw may not be structural equivalents because a nail employs a cylindrical surface to fasten wooden parts together, while a screw employs a helical surface, in the context of fastening wooden parts, a nail and a screw may be equivalent structures.
Claims
1. A method comprising: receiving data for downhole pump operation using an apparatus including a pump; analyzing the data using a local edge framework that executes a trained machine learning model to identify operating conditions of the downhole pump operation that are associated with a risk of shortening the useful life of the pump; as well as Instructions are issued to the device to resolve the operating condition.
2. The method of claim 1, wherein the operating conditions involve solids. The method of claim 2 , wherein the instructions include solids flushing process instructions. 4 . The method of claim 3 , wherein the instructions include instructions for controlling the speed of an electric motor operably coupled to the pump.
5. The method of claim 1, wherein the operating conditions involve a gas. The method of claim 5 , wherein the instructions include gas release process instructions.
7. The method of claim 6, wherein the instructions include instructions for controlling the speed of an electric motor operably coupled to the pump. The method of claim 1 , wherein the analyzing characterizes torque.
9. The method of claim 8, wherein the analyzing identifies torque peaks associated with solids accumulation.
10. The method of claim 8, wherein the analyzing identifies torque peaks associated with gas ingestion.
11. The method of claim 1 , wherein the trained machine learning model is trained via unsupervised learning.
12. The method of claim 1, wherein the trained machine learning model is trained via supervised learning.
13. The method of claim 1, wherein the trained machine learning model includes a threshold. The method of claim 13 , wherein the threshold comprises at least one dynamic threshold.
15. The method of claim 1, wherein the command comprises a pump speed control command.
16. The method of claim 1, wherein the data comprises gas turbine generator data.
17. The method of claim 16, wherein the gas turbine generator data corresponds to operation of a gas turbine generator that generates electrical power to operate at least the pump.
18. The method of claim 1, comprising coordinating issuance of control commands for a plurality of said downhole pump operations.
19. A system comprising: processor; a memory accessible to the processor; processor-executable instructions stored in the memory and executable by the processor to instruct the system to: receiving data for downhole pump operation using an apparatus including a pump; analyzing the data using a local edge framework that executes a trained machine learning model to identify operating conditions of the downhole pump operation that are associated with a risk of shortening the useful life of the pump; as well as Instructions are issued to the device to resolve the operating condition.
20. One or more computer-readable storage media comprising processor-executable instructions to instruct a computing system to: receiving data for downhole pump operation using an apparatus including a pump; analyzing the data using a local edge framework that executes a trained machine learning model to identify operating conditions of the downhole pump operation that are associated with a risk of shortening the useful life of the pump; and Instructions are issued to the device to resolve the operating condition.
