Systems and methods for generating performance insights in well construction processes
A machine learning-based system analyzes historical well construction data to optimize parameters, addressing inefficiencies by identifying influential variable combinations and providing real-time recommendations for improved efficiency and cost-effectiveness in well construction processes.
Patent Information
- Application Number
- PCT/US2026/027298
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- Priority Date
- 2025-05-13
- Filing Date
- 2026-05-08
- Publication Date
- 2026-11-19
AI Technical Summary
Existing well construction processes face challenges in efficiently adjusting parameters due to the large number of variables involved, making it difficult to optimize time, efficiency, and integrity, with existing methods failing to account for the complex interactions between input variables.
A data processing system employs a machine learning algorithm to analyze historical well construction data, identifying influential combinations of variables and generating visualizations and real-time recommendations to optimize well construction parameters, using sensors and equipment data to improve efficiency and reduce costs.
The system provides insights and recommendations that enhance well construction efficiency by identifying key parameters for time and cost savings, allowing for real-time adjustments to improve the well construction process.
Smart Images

Figure US2026027298_19112026_PF_FP_ABST
Abstract
Description
SYSTEMS AND METHODS FOR GENERATING PERFORMANCE INSIGHTS IN WELL CONSTRUCTION PROCESSESCROSS REFERENCE
[0001] This application claims the benefit of and priority to U.S. Non-Provisional Application No. 19 / 206,432, filed May 13, 2025, the disclosure of which is hereby incorporated herein by reference.BACKGROUND
[0002] The present disclosure generally related generating well construction plans. More specifically, the present disclosure is generally related to identifying suitable well construction operational plans and procedures that expedite and improve the well construction process.
[0003] Wells are generally drilled into the ground or ocean bed to recover natural deposits of hydrocarbon deposits, such as oil, gas, and other materials that may be trapped in subterranean formations. Well construction operations (e.g., drilling operations) may be performed at a wellsite by a well construction system (e.g., drilling rig) having various surface and subterranean well construction equipment being operated in a coordinated manner. For example, a drive mechanism, such as a top drive located at a wellsite surface, can be utilized to rotate and advance a drill string into a subterranean formation to drill a wellbore.
[0004] Well construction operators may monitor and analyze a wide breadth of information to ensure that the well is being constructed in an efficient and safe manner. However, in view of the large amount of variables associated with the well construction process, it may be difficult to adjust well construction process parameters to account for the different scenarios that may be encountered while constructing a well. As such, improvements to generating well construction instructions, plans, and schedules may enable well to be constructed in a more efficient manner.
[0005] This section is intended to introduce the reader to various aspects of art that may be related to various aspects of the present techniques, which are described and / or claimed below. This discussion is believed to be helpful in providing the reader with background information to facilitate a better understanding of the various aspects of the present disclosure. Accordingly, it should be understood that these statement are to be read in this light, and not as admission of prior art.SUMMARY
[0006] A summary of certain embodiments disclosed herein is set forth below. It should be understood that these aspects are presented merely to provide the reader with a brief summary of these certain embodiments and that these aspects are not intended to limit the scope of this disclosure. Indeed, this disclosure may encompass a variety of aspects that may not be set forth below.
[0007] In one embodiment, A method may include receiving well construction process data that may include a plurality of datasets associated with constructing a plurality of wells over a period of time and receiving well completion data associated with the well construction process data. The well completion data may include output parameters associated with constructing the plurality of wells. The method may then involve generating a well completion model based on the well construction process data and the well completion data, such that the well completion model includes well insight data associated with impacts of at least a portion of the plurality of datasets with respect to at least a portion of the plurality of output parameters. The method may involve generating visualizations based on the well insight data, receiving an input at the visualizations, and sending commands to equipment associated with a well construction project in response to receiving the input.
[0008] The brief summary presented above is intended only to familiarize the reader with certain aspects and contexts of embodiments of the present disclosure without limitation to the claimed subject matter.BRIEF DESCRIPTION OF THE DRAWINGS
[0009] These and other features, aspects, and advantages of the present disclosure will become better understood when the following detailed description is read with reference to the accompanying drawings in which like characters represent like parts throughout the drawings, wherein:
[0010] FIG. 1 illustrates an example well site that may be employed to drill a wellbore through subsurface formations, in accordance with an embodiment;
[0011] FIG. 2 illustrates a data flow diagram for processing well construction process data to estimate well completion data, in accordance with an embodiment;
[0012] FIG. 3 illustrates example visualizations associated with SHAP values for various well construction process parameters, in accordance with an embodiment;
[0013] FIG. 4 illustrates example visualizations associated with expected well completion insights, in accordance with an embodiment;
[0014] FIG. 5 illustrates example visualizations associated with expected well completion impacts associated with certain well construction process parameters, in accordance with an embodiment; and
[0015] FIG. 6 illustrates a flow diagram of a method for processing well construction process data to generate estimated well completion data, in accordance with an embodiment;DETAILED DESCRIPTION
[0016] Certain embodiments commensurate in scope with the present disclosure are summarized below. These embodiments are not intended to limit the scope of the disclosure, but rather these embodiments are intended only to provide a brief summary of certain disclosed embodiments. Indeed, the present disclosure may encompass a variety of forms that may be similar to or different from the embodiments set forth below.
[0017] As used herein, the term “coupled” or “coupled to” may indicate establishing either a direct or indirect connection (e.g., where the connection may not include or include intermediate or intervening components between those coupled), and is not limited to either unless expressly referenced as such. The term “set” may refer to one or more items. Wherever possible, like or identical reference numerals are used in the figures to identify common or the same elements. The figures are not necessarily to scale and certain features and certain views of the figures may be shown exaggerated in scale for purposes of clarification.
[0018] As used herein, the terms “inner” and “outer”; “up” and “down”; “upper” and “lower”; “upward” and “downward”; “above” and “below”; “inward” and “outward”; and other like terms as used herein refer to relative positions to one another and are not intended to denote a particular direction or spatial orientation. The terms “couple,” “coupled,” “connect,” “connection,” “connected,” “in connection with,” and “connecting” refer to “in direct connection with” or “in connection with via one or more intermediate elements or members.”
[0019] Furthermore, when introducing elements of various embodiments of the present disclosure, the articles “a,” “an,” and “the” are intended to mean that there are one or more of the elements. The terms “comprising,” “including,” and “having” are intended to be inclusive and mean that there may be additional elements other than the listed elements. Additionally, it should be understood that references to “one embodiment,” “an embodiment,” or “some embodiments” of the present disclosure are not intended to be interpreted as excluding the existence of additional embodiments that also incorporate the recited features. Furthermore, the phrase A “based on” B is intended to mean that A is at least partially based on B. Moreover, unless expressly stated otherwise, the term “or” is intended to be inclusive (e.g., logical OR) and not exclusive (e.g., logical XOR). In other words, the phrase A “or” B is intended to mean A, B, or both A and B.
[0020] The present disclosure is directed towards improving systems and methods related to determining and modifying parameters (e.g., well construction process parameters) related to a well construction process. A well construction process may be related to a set ofcoordinated operations performed by personnel and equipment to facilitate the construction of a well within the earth. With this in mind, a number of variables related to the well construction process may affect the time, efficiency, and integrity in which the well may be constructed. By way of example, the variables may include information related to a work crew associated with performing the well construction process, a type of drilling rig employed for the well construction process, a type of tube employed in the well, and the like. To determine the well construction process parameters to employ when drilling a well, a data processing system may estimate certain completion parameters (e.g., drilling connection time, casing moving time, casing connecting time, etc.) associated with drilling the well by individually analyzing certain input variables (e.g., crew type, rig, motor type, drill size, stabilizer type, etc.) with respect to its effects to the completion parameters.
[0021] Although by individually analyzing each of the input variables, a data processing system may identify some input variables that may individually influence the completion parameters more than other input variables, the relationship between different combinations of input variables may not be accounted for when calculating the effectiveness. That is, the combination of certain input variables may be related to a more impactful effect in the completion parameters that may not be identifiable when evaluating input variables individually. Moreover, individually analyzing each input variable may be an inefficient process with respect to processing time, memory utilization, computing power consumption, and the like. With this in mind, to reduce an amount of time and resources associated with the evaluation of certain input variables related to the well construction process, the data processing system may identify the input variables that may be more influential than other input variables using a machine learning algorithm that leverages historical data associated with a number of other well construction processes.
[0022] The historical data may include input variables and well construction process data acquired during other well construction processes in the past. The process data may include sensor data acquired from sensors on a drilling rig of a well, drilling parameter operations (e.g., speed, torque, rate of penetration (ROP)) of the drilling rig, subsurface data (e.g., formation, rock type), personnel data (e.g., crew names), equipment data (e.g., drilltechnology), and the like. In some embodiments, the data processing system may employ the machine learning algorithm to determine relationships between the process data and the completion parameters (e.g., time, costs) related to well construction. Based on the identified relationships, the data processing system may identify a set or collection of variables that may improve the completion parameters with respect to time, energy consumption, cost, and the like. For example, the data processing system may determine a relationship between well construction process parameters, such as a rate of penetration for drilling a well, a drill bit size, a motor type (e.g., rotary steerable system (RSS)), a crew, and the like, with respect to the completion parameters. The data processing system may use the relationships to identify the well construction process parameters that may be more effective in drilling wells in a particular area.
[0023] By way of example, the machine learning algorithm may enable the data processing system to identify the combination or collection of well construction process parameters that may be related to providing the largest amount of energy, time, and cost savings associated with the well construction process. Based on these identified parameters, the data processing system may determine certain insights (e.g., expected effects or impacts to parts of the well construction process) related to various aspects of the well construction process and may generate visualizations to present the insights to allow a user to review the insights.
[0024] In some embodiments, the data processing system may generate visualizations to present certain operations of the well construction process associated with an expected impact value (e.g., time saved, cost saved, etc.). The data processing system may determine the expected impact value of each operation based on the relationships and evaluations of associated drilling parameter operations and identified variables. For example, the data processing system may determine that reducing a time for a drilling connection operation may be correlated with a particular crew and tubular type of a drilling tool. The data processing system may then quantify an expected impact value with respect to certain measurements, such as minutes that may represent a potential time saved for the drilling connection operation.
[0025] In some embodiments, the data processing system may generate recommendations to a user that include suggested operational procedures (e g. identified well construction process parameters) of the operation. The recommendation may also include visualizations of relationships between completion operations and the identified well construction process parameters. For example, the data processing system may then recommend the user to employ a certain crew and tubular type of a drilling tool to reduce time associated with performing a drilling connection. With this in mind, the data processing system may generate visualizations that demonstrates relationships between a drilling connection time and the crew and tubular type of the drilling tool based on the insights determined using the machine learning algorithm.
[0026] In some embodiments, the data processing system may also be used in a real time well construction process. As such, for example, the data processing system may receive real time well construction process data that include drilling parameter operations and variable data of a rig. The data processing system may employ a well construction process model that may be generated based on the insights determined using the machine learning algorithm and the historical data. The well construction process model may use the real time data to identify well construction process parameters that may be adjusted to affect a current operation being performed on the rig. The data processing system may then recommend the user or operators on the rig to change current operational procedures to improve performance of the current operation with respect to time and costs savings as indicated by the well construction process model.
[0027] For example, the well construction process model may indicate that certain stabilizer types and bit types of a drilling tool may be more impactful for a drilling operation of a certain rig as compared to the stabilizer types and bit types of the drilling tool currently being employed according to the real time data received by the data processing system. Based on this insight, the data processing system may generate a notification to send to computing devices of operators that recommend to change the current stabilizer type and bit type of the drilling tool. In the same manner, in some embodiments, the well construction process model may indicate drilling parameters such as speed, torque, and the like to adjust based on thereal time data acquired from sensors within the rig and the well construction process model. In turn, the data processing system may generate commands to drive systems to modify their respective operations to achieve the drilling parameters determined using the well construction process model. By modifying the drilling parameters in real time while performing the well construction process, the data processing system may increase the efficiency in construction the well.
[0028] By employing a machine learning algorithm and accounting for drilling parameter operations and variables related to operations performed on a rig, the present embodiments provide an improved methodology for identifying well construction process parameters that may impact the operations of a well construction process in a manner that provides a cost-effective design. Moreover, by leveraging historical data using the machine learning algorithm, the present embodiments may enable computing systems to quantify an impact (e.g., time saved, cost saved, etc.) for the operations of the well construction process in a more time and cost-efficient manner. Additional details regarding the present embodiments will be discussed below with reference to FIGS. 1-6.
[0029] By way of introduction, FIG. 1 shows an example well construction site 10 configuration for acquiring well data that may use, among other sensing devices, a logging while drilling (LWD) system 12. The LWD system 12 may include one or more collar sections 14, 16, 18, 38 coupled to the lower end of a drill pipe 22. The LWD system 12 may form part of a bottom hole assembly (BHA) coupled to the drill pipe 22 and includes at a lower end thereof a drill bit 24 to drill the wellbore 26 through the earth's subsurface 28. Drilling is performed by rotating the drill pipe 22 by means of a rotary table 30, or in other examples by a top drive (not shown in FIG. 1). In still other implementations a drilling motor (not shown) may be used to rotate some or all of the drill pipe 22. The manner of rotating the drill pipe 22 is not a limit on the scope of the present disclosure. During rotation, the drill pipe 22 is suspended by equipment on a drilling equipment (e.g., drilling rig) including a swivel 32 which enables the drill pipe 22 to rotate while maintaining a fluid tight seal between the interior and exterior of the drill pipe 22. Mud pumps 34 draw drilling fluid ("mud") 36 from a tank or pit and pump the mud 36 through the interior of the drill pipe 22, down throughthe LWD system 12, as indicated by arrow 38. The mud 36 passes through orifices (not shown) in the drill bit 24 to lubricate and cool the bit 24, and to lift drill cuttings in through an annulus 40 between the drill pipe 22, LWD system 12 and the wellbore 26. The cuttings may be collected and analyzed by separate equipment operated by a "mud logger." Any fluid content in the returned drilling mud, such as gas and / or oil may also be analyzed by the mud logger equipment or other devices, such as an automatic gas chromatograph. The mud logger may provide information related to analysis of cuttings that may be analyzed by a data processing system or the like.
[0030] The collar sections 14, 16, 18, 20 may include sensors (not shown) therein which make measurements of various properties of the subsurface earth formations 28 through which the wellbore 26 is drilled. These measurements are typically recorded in a recording device (not shown) disposed in one or more of the collar sections 14, 16, 18, 20. LWD systems known in the art typically include one or more "measurement while drilling" (MWD) sensors (not shown separately), which measure selected drilling parameters, such as inclination and azimuthal trajectory of the wellbore 26. Other drilling sensors known in the art may include axial force (weight) applied to the LWD system 12, shock and vibration sensors, and the like.
[0031] In some embodiments, the LWD system 12 may include a mud pressure modulator (not shown separately) in one of the collar sections 16. The mud pressure modulator (not shown) applies a telemetry signal to the flow of mud 36 inside the LWD system 12 and drill pipe 22 where it may be detected by a pressure sensor 42 disposed in the mud flow system. The pressure sensor 42 may be coupled to detection equipment (not shown) in a data processing system 44, which enables recovery and recording of information transmitted in the telemetry scheme sent by the LWD system 12. The telemetry scheme may include a subset of measurements made by the various sensors (not shown separately) in the LWD system 12. The remainder of the measurements made by the sensors (not shown) in the system may be transferred, for example, using radio telemetry or a cable connection, to the data processing system 44 when the LWD system 12 is withdrawn from the wellbore 26.
[0032] The well construction site 10 shown in FIG. 1 is merely provided to serve as an example of how the wellbore 26 may be constructed and data are acquired at the well site, and is not in any way intended to limit the scope of the disclosure. Other sources of data may include control systems for wellbore pressure control. See, for example, U.S. Pat. No.6,904,981 issued to van Riet. The system described in the van Riet '981 patent can provide automatic control over wellbore fluid pressure, and may also calculate parameters such as expected formation fluid pressure and expected formation fracture pressure. Such data may also be communicated as will be further explained below. Still other sources of data may include, without limitation, so-called "mudlogging" data as explained above, wherein drilling fluid returned from the wellbore is analyzed for the presence of materials such as hydrocarbons, and samples of drill cuttings are analyzed for mineral content and grain structure. Still other data may include casing programs (i.e., depth to which casings are set and respective diameters thereof and types of cement to be used) and planned wellbore geodetic trajectory. Any one or more of the foregoing data types, whether measured during drilling of the wellbore, entered into a computer system (explained below) manually or otherwise, may be referred to as a "wellbore construction parameter." The drilling rig operator (driller) may operate the drilling unit, typically by manual input to a control panel of a programmable logic controller (PLC-not shown in FIG. 1) that operates various devices on the drilling rig such as the rate at which the pump 34 discharges drilling fluid, the rate at which the swivel 32 (or top drive) is lowered into the wellbore, thus affecting the axial force (weight) on the drill bit 24, which may be inferred by a load cell 25 that measures the axial loading on the drill pipe 22, and the rotating speed of the drill bit 24. The foregoing parameters, which are controllable by the drilling unit operator, may be referred to as "drilling operating parameters." Another sensor that may be used in some examples is a torque sensor 46 that may be coupled to the rotary table 30 if used, or may be included in the top drive (if used; not shown in FIG. 1).
[0033] In FIG. 1, the data processing system 44 may include a data communication subsystem 48 coupled to a suitable antenna, satellite dish or other communication link 50. Such data communication subsystem 48 may be of any type known in the art suitable for use at the particular location of the well site, for example, satellite communication to the Internet,or a dedicated satellite based communication link. Radio communication, wired communication or any other form of data communication is within the scope of the communication subsystem 48 applicable to the present example method and system and the foregoing examples should not be considered limiting. Communication may take place over any form of data network.
[0034] For purposes of the present disclosure, the drilling equipment and operations shown in FIG. 1 may be merely examples of equipment that may be monitored by a well construction engineer or drilling engineer from any location and using any device that can access a computer system. Such access devices may be directly connected to the computer system, or may be remotely accessible, such as by wireless broadband systems using Internet-capable communication (e.g., tablets or smart phones), dedicated-channel wireless or radio communication terminals, or telephone network connected terminals (e.g., DSL Internet access). Monitoring and advising operations on a plurality of well construction projects will be further explained below.
[0035] The data processing system 44 may be any electronic data processing system that can be used to carry out the systems and methods of this disclosure. For example, the data processing system 44 may include one or more processors 52, which may execute instructions stored in memory 54 and / or storage 56. The memory 54 and / or the storage 56 of the data processing system 44 may be any suitable article of manufacture that can store the instructions. In certain embodiments, the one or more processors 52 may include a microprocessor, a microcontroller, a processor module or subsystem, a programmable integrated circuit, a programmable gate array, a digital signal processor (DSP), or another control or computing device. In certain embodiments, the one or more processors 52 may include machine learning and / or artificial intelligence (Al) based processors.
[0036] In certain embodiments, the memory 54 and storage 56 may be implemented as one or more non-transitory computer-readable or machine-readable storage media. In certain embodiments, the memory 54 may include one or more different forms of memory, including semiconductor memory devices such as dynamic or static random-access memories (DRAMs or SRAMs), erasable and programmable read-only memories (EPROMs), electricallyerasable and programmable read-only memories (EEPROMs) and flash memories. The storage 56 may include solid state drives, magnetic disks such as fixed, floppy and removable disks; other magnetic media including tape; optical media such as compact disks (CDs) or digital video disks (DVDs); or other types of storage devices. Note that the computerexecutable instructions and associated data of the analysis module(s) may be provided on one computer-readable or machine-readable storage medium of the memory 54 or the storage 56, or alternatively, may be provided on multiple computer-readable or machine-readable storage media distributed in a large system having possibly plural nodes. Such computer-readable or machine-readable storage medium or media are considered to be part of an article (or article of manufacture), which may refer to any manufactured single component or multiple components. In certain embodiments, the storage 56 may be located either in the machine running the machine-readable instructions or may be located at a remote site from which machine-readable instructions may be downloaded over a network for execution.
[0037] As illustrated, the data processing system 44 may optionally also include a display 58, which may be any suitable electronic display, and may display images generated by the processor 52. The data processing system 44 may be a local component of the well site 10 (i.e., at the surface), within the LWD system 12 (i.e., downhole), a device located proximate to the drilling operation, and / or a remote data processing device located away from the well site 10 to process downhole measurements in real time or sometime after the data has been collected. In some embodiments, the data processing system 44 may be a portable computing device (e.g., tablet, smart phone, or laptop) or a server remote from the well site 10. In some embodiments, the LWD system 12 may store and process collected data in the bit 24 (e.g. bottom hole assembly) or send the data to the surface for processing via the data communication system 48 described above, including any suitable telemetry.
[0038] It should be noted that, although the discussion above relates to a well site, other downhole equipment or systems may employ the systems and methods of this disclosure. For example, a downhole tool with an acoustic tool conveyed by slickline, coiled tubing, wireline, or other delivery systems, may utilize the disclosed systems and methods.
[0039] Keeping the foregoing in mind, FIG. 2 illustrates a data flow chart 70 of example datasets that may be provided to the data processing system 44, such that the data processing system 44 may generate various outputs or insights associated with the analysis of the input datasets. Although the following description of the data flow chart 70 includes certain input datasets and certain output datasets, it should be understood that these datasets are provided as examples and the embodiments described herein may include fewer or additional suitable datasets.
[0040] By way of operation, in some embodiments, the data processing system 44 may receive a variety of datasets associated with well construction process data. The well construction process data may be representative of certain aspects with respect to constructing a well within the subsurface region of the earth. As such, in some embodiments, the well construction process data may include drilling data 72 that may be available from sensors or various data sources (e.g., LWD system 12) as described above in FIG. 1. Indeed, the drilling data 72 may include inclination and azimuth data related to the borehole, toolface orientation data related to steering the drill equipment, drilling dynamic data (e.g., weight on bit, torque on bit, rate of penetration, rotary speed (revolutions per minute (RPM), vibration data (e.g., axial, lateral, torsional), bounce data), mud data (e.g., downhole pressure, temperature, mud flow rate, equivalent circulating density (ECD)), and the like. That is, the drilling data 72 may include data related to the drilling of the borehole for the construction of the well. The drilling data 72 may be received via sensors disposed on the drilling rig, within the borehole, and the like. In addition, the drilling data 72 may be received via drilling equipment performing the drilling operations, such as drive system for operating a motor.
[0041] In addition to the above, the drilling data may include real-time drilling data, such as surface drilling parameters (e.g., hookload, pressure, pump rate), bit depth, hole depth, bottom hole assembly (BHA) details (e.g., components, configuration, stabilization), drilling fluid properties (e.g., density, viscosity, pH, solids content, filtrate loss), gas detection, and the like. In some embodiments, the LWD system 12 may provide gamma ray data, resistivity data, density data, and neutron data related to the subsurface region being drilled.
[0042] Based on the drilling data 72, the data processing system 44 may determine various aspects of the drilling operation. That is, the drilling data 72 may enable the data processing system 44 to determine that the well construction process is at a drilling stage, a casing stage, a connection stage, or the like. In some embodiments, the data processing system 44 may receive contextual data 74 that may provide context or descriptions related to the stages of the well construction process relative to the drilling data 72 or other input datasets. The contextual data 74 may be provided by users (e.g., well operators) in text form to reporting systems and the like. For example, the users may provide text data to reporting systems (e.g., software) during the course of the well construction process. The text data may indicate certain tasks that have been completed, times that the tasks have been completed, personnel or crews that performed certain tasks, and the like.
[0043] In some embodiments, the contextual data 74 may be received or detected via messages provided by the users to the reporting system or to other individuals. The messages may include electronic mail messages, instant messages via chat applications, or the like. The times associated with the messages may be mapped to the drilling data 72 to determine the context or description of the corresponding stage of the well construction process being performed, completed, or the like. In some embodiments, the data processing system 44 may employ a large language model (LLM) to gather summaries and generate the contextual data 74 based on correlations or relationships identified between the text data and the drilling data 72.
[0044] In addition to the drilling data 72, the data processing system 44 may receive casing data 76, which may include information related to the casing and / or cementing associated with the well being constructed. The casing data may include casing setting depths, diameters, grades, and the like. The cement data may include cement slurry properties, volumes, displacement rates, and the like.
[0045] The workforce data 78 may include information related to the crews or personnel that performs different tasks or operations of the well construction process. As such, the workforce data 78 may include a crew name to identify a group of individuals associated witha particular crew, times that the crew performed certain tasks, cost data for using the respective crew, and the like.
[0046] In some embodiments, the data processing system 44 may receive subsurface data 80 that may include representations of the rock formations, hydrocarbon deposits, and the like within the subterranean region. As such, the subsurface data 80 may include seismic data generated during seismic surveys. In addition, the subsurface data 80 may include wireline logs that may include sonic data, nuclear magnetic resonance (NMR) data, acoustic data, nuclear data, formation pressure testing data, and the like. The subsurface data 80 may also include geological models that provide stratigraphy, structure, and reservoir representations.
[0047] The data processing system 44 may also receive environmental data 82. The environmental data 82 may include weather data that provides temperature, pressure, humidity, wind, storm, and other relevant information regarding the weather at various times with respect to well site locations. The environmental data 82 may also coordinate data associated with various well sites. In this way, the data processing system 44 may have elevation data and location data associated with the well site.
[0048] In some embodiments, the data processing system 44 may also other completion data 84 in addition to the casing data 76. For example, the completion data 84 may include perforation parameters including perforation intervals / depths, shot density data, phasing / orientation data, and the like. The completion data 84 may also include completion hardware data such as packer depth, sliding sleeve data, flow control device locations, and the like. In some embodiments, the completion data 84 may provide information regarding the dates or times in which various stages of the well construction process were completed. In addition, the completion data 84 may provide duration of times associated with performing these various stages or parts of the well construction process. The stages of the well construction process may include a connection stage, a pre-connection stage, a postconnection stage, a moving time, a connection time, a casing moving time, a casing connection time, pick up times for BHAs, casing, or tubing, lay-down times for tubing, BHA, and joints, and the like.
[0049] In addition to the completion data 84, the data processing system 44 may also receive production data 86 related to the flow or production of hydrocarbons from the well. The production data 86 may include production logging (PLT), well integrity logs, reservoir monitoring, flow rates, and the like. The production data 86 may also include well test data, artificial lift parameters, and fluid properties. The production data 86 may be received via sensors within the well or based on calculations determined using datasets associated with the well.
[0050] Each of the datasets described above may correspond to historical datasets associated with one or more well sites in one or more different parts of the world. As such, the data processing system 44 may collect datasets related to a variety of geological regions, a variety of well types, a variety of equipment employed to construct the wells, a variety of crews working to construct the wells, and the like. Based on this historical data, the data processing system 44 may apply certain machine learning algorithms to predict completion data for other well sites. The machine learning algorithms may include any suitable algorithm that may identify trends or patterns with respect to the input datasets to make predictions for other similar datasets. The machine learning algorithms may include supervised learning algorithms, such as linear regression, ridge / Lasso regression, decision tree regressor, random forest regressor, and the like. The machine learning algorithms may also include unsupervised learning algorithms, such as clustering, K-means, principal component analysis, and the like. In any case, the data processing system 44 may apply the machine learning algorithms to the input datasets to predict completion parameters (e.g., completion data) for other well construction processes.
[0051] In some embodiments, the data processing system 44 may identify trends, patterns, and / or correlations within the drilling data 72, the contextual data 74, the workforce data 78, the subsurface data 80, the environmental data 82, and the like with respect to the casing data 76, the completion data 84, production data 86, and the like. That is, the data processing system 44 may predict certain parameters as provided in the casing data 76, the completion data 84, production data 86, and the like based on the patterns identified by the machine learning algorithms with respect to the drilling data 72, the contextual data 74, theworkforce data 78, the subsurface data 80, the environmental data 82, and the like. In some embodiments, the data processing system 44 may represent the predictions and insights determined by the machine learning algorithms in a well completion model 88.
[0052] The well completion model 88 may provide predictions or estimates for various completion aspects of a well construction process relative to certain well construction process data (e.g., input datasets). As such, the well completion model 88 may provide estimated completion data with regard to a well to be constructed in the future. In this way, the data processing system 44 may provide input datasets representative of the well construction process data to be used for a planned well construction project to determined estimated completion data, such as time, costs, energy used (e.g., sustainability), and the like, associated with the well construction.
[0053] It should be noted that the well completion model 88 is generated based on the variety of input datasets and thus may identify patterns and correlations related to a collection of different input datasets and their collective effects to the completion data. That is, in contrast to evaluating each input dataset individually with respect to a parameter of the completion data, the present embodiments enable the data processing system 44 to evaluate different aspects of the identified trends to better ascertain insights and reasons for the changes in the completion data. For example, when evaluating input datasets individually, the data processing system 44 may determine that certain wells are associated with a higher rate of penetration (ROP) value as compared to others. However, the data processing system 44 may be limited in determining the reasons why the difference in the ROP exists. By leveraging the additional datasets provided to the data processing system 44, the data processing system 44 may identify other well construction process parameters that contribute to the improved ROP values, such as the speed of the drilling, the drill bit employed, the drilling equipment, the weather, and the like. In other words, by collecting the variety of datasets and employing the machine learning algorithms, the data processing system 44 may generated the well completion model 88 that provides increased dimensionality to evaluate different dimensions related to each input dataset that may contribute to the changes in the completion data.
[0054] With this in mind, in some embodiments, after determining the well completion model 88, the data processing system 44 may generate well completion insights 90 that characterize various properties or dimensions associated with some detected change. In some embodiments, the well completion insights 90 may include information related to a degree of impact or effect that any particular variable may have on an output or metric of the well completion model 88. That is, the machine learning algorithm used to generate the well completion model 88 may provide information with regard to the input datasets or variables that influenced the outcomes of the well completion model 88.
[0055] By way of example, the data processing system 44 may determine a Shapely Additive Explanation (SHAP) value associated with different input datasets to indicate a value that corresponds to an amount or degree of contribution of the input dataset with respect to any particular prediction or metric output by the well completion model 88. FIG. 3 illustrates an example chart of SHAP values that represent an impact to an output of the well completion model 88. For instance, as shown in FIG. 3, “Night Crew D” (e.g., workforce data) is associated with higher SHAP values as compared to the motor type used. In this way, the data processing system 44 may provide insight into the impact or weight of any particular input variable with respect to a model output.
[0056] In some embodiments, the well completion insights 90 may be used to generate well completion visualization 92 that may represent some of the information gleaned from analysis of the well completion model 88. For example, FIG. 4 illustrates example insight visualizations generated for a graphical user interface (GUI) 100 that may include information related to the well completion insights 90. For example, the data processing system 44 may use the output datasets or outputs of the well completion model 88 to identify the output parameters (e.g., completion data) that have a relatively large variability in values for a relatively large amount of time as compared to other outputs. After identifying these output datasets that have a wide range of values and are associated with a large amount of time of the well construction process relative to other output datasets, the data processing system 44 may generate visualization 102 to represent these characteristics. That is, for example, the pre-connection parameter may have a wide variability with respect to time.
[0057] The GUI 100 may include interactive objects that may allow a user to select one of the output parameters for further analysis, which may also correspond to the well completion insights 90. For instance, if the user wishes to learn more about the parameters that affect the pre-connection output, the GUI 100 may receive an indication of a selection of a portion of the visualization 102 that corresponds to the pre-connection. After receiving the selection, the data processing system 44 may generate a SHAP value chart 104 that illustrates different parameters that may be related to the variability of the pre-connection output. That is, the pre-connection output may indicate that the time associated with performing the preconnection may vary by a wide margin as compared to the connection time and on bottom drilling time, as indicated in the visualization 102. To better understand the parameters or factors that affect the variability of the pre-connection output, the data processing system 44 may identify the input datasets or parameters that have the most impact or weight on the change to the pre-connection time by generating the SHAP value chart 104 or performing some other impact analysis operation.
[0058] After determining the impacts associated with each input variable relative to some output parameter, the data processing system 44 may generate visualizations that illustrate the impacts in units corresponding to the output. For example, FIG. 5 illustrates impacts in units of minutes for different completion parameters, such as on bottom drilling operation, drilling connection operation, and circulating operation. That is, the visualization depicted in FIG. 5 may correspond to the impacts to different completion parameters that may be associated with different input datasets or parameters. In this way, the data processing system 44 or a user may better understand the benefits or detriments related to adjusting certain input parameters.
[0059] In some embodiments, the GUI 100 may include additional visualizations to illustrate other output data or completion data, such as average ROP for different wells, as presented in visualization 106. It should be noted that a variety of completion data may be presented via a visualization based on the well completion insights 90 or other suitable data.
[0060] Referring back to FIG. 2, based on the well completion insights 90, the impact values, and other analysis performed by the data processing system 44, the data processingsystem 44 may generate well completion recommendations 94. The well completion recommendations 94 may provide instructions related to achieving improved completion data or parameters thereof. In some embodiments, the data processing system 44 may receive an input requesting recommendations for improving some completion parameter, such as preconnection time.
[0061] After identifying the well construction process parameters that may provide the highest relative impact or weight to the respective completion parameter, the data processing system 44 may generate the well completion recommendations 94 to assist a user to achieve improved completion parameters. In some embodiments, the data processing system 44 may leverage the context data 74 and a large language model to generate instructions that provide details for achieving the improved completion parameters. Indeed, the data processing system 44 may identify the input parameters that may be adjusted to provide improved completion parameters based on the impact analysis described above. The data processing system 44 may then use the large language model to generate the instructions as text, simulated audio, or the like, such that the user may review, confirm, and implement the instructions. For instance, as depicted in FIG. 4, a visualization 108 may provide insight recommendations that provide information related to the well completion insights 90 or the well completion recommendations 94.
[0062] In some embodiments, the large language model may be used to generate commands or instructions for devices to adjust their respective operations. For instance, the instructions may be related to adjusting the speed of a motor of drill equipment at certain depths of the borehole, and the data processing system 44 may send the instructions to the drilling equipment to modify its operations in accordance with the instructions.
[0063] With the foregoing in mind, FIG. 6 illustrates an example flow chart of a method 120 for generating well completion insights, in accordance with embodiments described herein. Although the following description of the method 120 will be discussed as being performed by the data processing system 44 and in a particular order, it should noted that the method 120 may be performed in any suitable order by any suitable computing device.
[0064] Referring now to FIG. 6, at block 122, the data processing system 44 may receive a selection of well construction process data or parameters via an input device, a graphical user interface, or the like. The selection may be provided by a user that may be analyzing different well construction process parameters and different combinations of the well construction process parameters for constructing a well. As discussed above, the well construction process parameters may include information related to different aspects or stages associated with constructing a well. As such, the well construction process data may include the drilling data 72, the workforce data 78, the subsurface data 80, the environmental data 82, and the like. In some embodiments, the well construction process data may be associated with a number of wells constructed across a number of well sites in different regions of the world. As such, each dataset of the well construction process data may provide a reference or certain expected features / characteristics for constructing a well in the respective region.
[0065] At block 124, the data processing system 44 may receive context data associated with the well construction process data received at block 122. The context data may correspond to the contextual data 74 described above and may provide user inputs or descriptions of different aspects of the well construction process data. That is, the context data may provide descriptions with respect to certain identified anomalies in sensor readings, drill equipment functions, and the like. The context data may provide a text-based description of various aspects of the well construction process data, information related to different stages of various well construction processes, information related to time period between the stages, and the like. In some embodiments, the data processing system 44 may provide the context data to a large language model as training data to enable the large language model to more efficiently identify context for corresponding well construction process datasets.
[0066] At block 126, the data processing system 44 may receive completion data. The completion data may correspond to outputs, measured responses, and other datasets that correspond to milestones or completion of stages of the well construction process. The completion data may include the casing data 76, the completion data 84, the production data 86, and the like. As such, the data processing system 44 may associate the well constructionprocess data with corresponding observed outputs, such as a casing depth, time periods for performing certain tasks, hydrocarbon production rates, and the like.
[0067] Based on the datasets received at blocks 122, 124, and 126, the data processing system 44 may, at block 128, generate a well completion model, as described above. The well completion model may be generated using certain machine learning algorithms that identify correlations between the well construction process data and the completion data received at blocks 122 and 126, respectively. That is, the well completion model may evaluate the combination of multiple variables of the well construction process data and their collective effects to the completion data. Indeed, while other analysis systems individually evaluate the effect of each individual dataset of the well construction process data with respect to an individual well completion parameter, the present embodiments described herein analyze the well construction process data in a more comprehensive manner to identify the collective effect of different well construction process parameters on a respective well completion parameter.
[0068] In some embodiments, the well completion model may be generated based on the expected impacts of each respective parameter of the well construction process parameters. That is, the well completion model may account for the impact values for different well construction process parameters to provide weighted data related to certain effects to different completion parameters. In this way, the well completion model may include the impact or weight values to allow others to identify adjustments or recommendations in performing well construction operations that may be more impactful to certain completion parameters. For example, the impact value with respect to time for completing a well may be higher for drilling speeds used when drilling the well as compared to using a particular crew for the well construction. Indeed, since the well completion model characterizes the collective impacts of multiple well construction process variables, the data processing system 44 may be better equipped to identify efficient well construction process parameters to employ.
[0069] After generating the well completion model, the data processing system 44 may store the well completion model in a storage component, such as a database, a cloudcomputing system, or the like. In this way, different users may use the well completion modelto analyze different well construction processes being evaluated. Indeed, in some embodiments, the well completion model may be utilized by drilling systems to dynamically modify certain drill equipment operational parameters in real time based on real-time data acquired by the LWD system 12 or other sensors.
[0070] At block 130, the data processing system 44 may receive a request to provide well completion insights based on certain input well data. The input well data may be specified by a user and may correspond to the well construction process data described above. That is, the user may provide some details related to some input parameters associated with a well being constructed, being considered for construction, and the like. In some embodiments, the well input data may include information related to the crew identifications that are available, the type of drilling equipment that is accessible, the subsurface region being considered, and the like.
[0071] Based on the input well data, the data processing system 44 may, at block 132, generate well completion insight data based on the well completion model and the input well data. That is, the data processing system 44 may query the well completion model based on the input well data to identify insights related to improving an operation or function that may be specified in the request. As such, the data processing system 44 may identify a number of well construction process datasets or input variables that may provide the largest impact or effect to various well completion parameters (e.g., time, flowrate, energy use) associated with the request. By way of example, the request may be related to identifying more time efficient processes in constructing a well and the input well data may include certain input datasets related to the available crews, the type of subsurface region, the available drilling equipment, the time of year, the region of the world, and the like. Based on these input datasets, the well completion model may be utilized to identify a collection of values or settings of various well construction process parameters that may be most influential in achieving the requested well completion parameter.
[0072] After identifying the well completion insight data (e.g., well completion insights 90), the data processing system 44 may, at block 134, generate well completion visualizations, such as those illustrated in FIGS. 5 and 6, based on the well completion insightdata. The well completion visualizations may include information related to different well construction process parameters and their respective effects to one or more well completion parameters. The well completion visualizations may be presented via a GUI, the display 58, or any other suitable display to allow a user to analyze and interact with the represented data. Indeed, the well completion visualizations may be interactive objects that may receive inputs (e.g., click, selection) that may cause the data processing system 44 to execute some action or code in response to detecting the inputs.
[0073] With this in mind, at block 136, the data processing system 44 may receive an input at one of the well completion visualizations. The input may be provided by a user requesting to view or review additional information related to the respective well completion visualization. In response to receiving an input to review additional information, the data processing system 44 may produce additional well completion visualizations that provide additional context, related datasets, or the like.
[0074] In some embodiments, the well completion visualizations may include the well completion recommendations 94 described above. As such, a user may review different recommendations or instructions to achieve a requested well completion parameter. In some embodiments, the recommendations may include operational parameters for equipment used in the well construction process. As such, in response to detecting an input associated with a recommended operational change, the data processing system 44 may proceed to block 138 and generate commands for the respective equipment and send the commands to the equipment at block 140. In turn, the equipment may modify its operation based on the received command. In this way, the present embodiments may be employed for real time modification of operations of various equipment associated with the well construction process.
[0075] The technical effect of the disclosed embodiments include providing an improved system to evaluate a number of well construction process parameters together with respect to the collective effects on certain well completion parameters. Indeed, instead of individually analyzing individual well construction process parameters and their respective effects to individual well completion parameters, the presently disclosed techniques analyze differentcombinations of well construction process parameters and their corresponding effects. In this way, computing devices operate more efficiently by avoiding the sequential processing of individual well construction process parameters. Instead, by leveraging the machine learning algorithms and the impact analysis techniques described herein, computing devices may identify well completion insights in a more efficient and accurate manner by accounting for the combination of well construction process parameters. Indeed, when evaluating parameters in an individual fashion, the contribution to the effectiveness of a respective parameter may be inaccurately reflected in view of neglecting the contribution of other parameters that are being implemented or occurring at the same time. By employing the techniques described herein, the parameters that may be more impactful to contributing to achieving some completion parameters may be more efficiently (e.g., time, computational energy) as compared to other methodologies.
[0076] The subject matter described in detail above may be defined by one or more clauses, as set forth below.
[0077] A method may include receiving well construction process data comprising a plurality of datasets associated with constructing a plurality of wells over a period of time; receiving well completion data associated with the well construction process data, wherein the well completion data comprises a plurality of output parameters associated with constructing the plurality of wells; generating a well completion model based on the well construction process data and the well completion data, wherein the well completion model comprises well insight data associated with one or more impacts of at least a portion of the plurality of datasets with respect to at least a portion of the plurality of output parameters; generating one or more visualizations based on the well insight data; receiving an input at the one or more visualizations; and sending one or more commands to equipment associated with a well construction project in response to receiving the input, wherein the one or more commands are configured to adjust one or more operations of the equipment.
[0078] The method of any preceding clause, wherein the well construction process data comprises drilling data, contextual data, workforce data, subsurface data, environmental data, or any combination thereof.
[0079] The method of any preceding clause, wherein the well completion data comprises a time duration for performing a well construction task.
[0080] The method of any preceding clause, wherein generating the well completion model comprises applying a machine learning algorithm to the well construction process data and the well completion data to identify one or more correlations between the at least a portion of the plurality of datasets with respect to the at least a portion of the plurality of output parameters.
[0081] The method of any preceding clause, wherein the one or more visualizations comprise an indication of the one or more impacts of the at least a portion of the plurality of datasets with respect to the at least a portion of the plurality of output parameters.
[0082] The method of any preceding clause, wherein the one or more impacts correspond to one or more amounts of time.
[0083] The method of any preceding clause, wherein generating the well completion model comprises: identifying the at least a portion of the plurality of datasets having a degree of variability greater than an additional portion of the plurality of datasets; determining one or more weights associated with the at least a portion of the plurality of datasets affecting with the at least a portion of the plurality of output parameters; and generating the well completion model based on the one or more weights.
[0084] A non-transitory computer-readable medium, comprising computer-executable instructions that, when executed, are configured to cause a processing system to perform operations comprising: receiving well construction process data comprising a plurality of datasets associated with constructing a plurality of wells over a period of time; receiving well completion data associated with the well construction process data, wherein the well completion data comprises a plurality of output parameters associated with constructing the plurality of wells; generating a well completion model based on the well construction process data and the well completion data, wherein the well completion model comprises well insight data associated with one or more impacts of at least a portion of the plurality of datasets withrespect to at least a portion of the plurality of output parameters; generating one or more visualizations based on the well insight data; receiving an input at the one or more visualizations; and sending one or more commands to equipment associated with a well construction project in response to receiving the input, wherein the one or more commands are configured to adjust one or more operations of the equipment.
[0085] The non-transitory computer-readable medium of any preceding clause, wherein the well construction process data comprises drilling data, contextual data, workforce data, subsurface data, environmental data, or any combination thereof.
[0086] The non-transitory computer-readable medium of any preceding clause, wherein the well completion data comprises a time duration for performing a well construction task.
[0087] The non-transitory computer-readable medium of any preceding clause, wherein the instructions to cause the processing system to generate the well completion model comprises additional instructions that cause the processing system to perform additional operations comprising applying a machine learning algorithm to the well construction process data and the well completion data to identify one or more correlations between the at least a portion of the plurality of datasets with respect to the at least a portion of the plurality of output parameters.
[0088] The non-transitory computer-readable medium of any preceding clause, wherein the one or more visualizations comprise an indication of the one or more impacts of the at least a portion of the plurality of datasets with respect to the at least a portion of the plurality of output parameters.
[0089] The non-transitory computer-readable medium of any preceding clause, wherein the one or more impacts correspond to one or more amounts of time.
[0090] The non-transitory computer-readable medium of any preceding clause, wherein the instructions to cause the processing system to generate the well completion model comprises additional instructions that cause the processing system to perform additional operations comprising: identifying the at least a portion of the plurality of datasets having adegree of variability greater than an additional portion of the plurality of datasets; determining one or more weights associated with the at least a portion of the plurality of datasets affecting with the at least a portion of the plurality of output parameters; and generating the well completion model based on the one or more weights.
[0091] A system may include a storage component comprising computer-executable instructions; and a processing system configured to computer-executable instructions to cause a processing system to perform operations comprising: receiving well construction process data comprising a plurality of datasets associated with constructing a plurality of wells over a period of time; receiving well completion data associated with the well construction process data, wherein the well completion data comprises a plurality of output parameters associated with constructing the plurality of wells; generating a well completion model based on the well construction process data and the well completion data, wherein the well completion model comprises well insight data associated with one or more impacts of at least a portion of the plurality of datasets with respect to at least a portion of the plurality of output parameters; generating one or more visualizations based on the well insight data; receiving an input at the one or more visualizations; and sending one or more commands to equipment associated with a well construction project in response to receiving the input, wherein the one or more commands are configured to adjust one or more operations of the equipment.
[0092] The system of any preceding clause, wherein the well construction process data comprises drilling data, contextual data, workforce data, subsurface data, environmental data, or any combination thereof.
[0093] The system of any preceding clause, wherein the well completion data comprises a time duration for performing a well construction task.
[0094] The system of any preceding clause, wherein the processing system is configured to generate the well completion model by applying a machine learning algorithm to the well construction process data and the well completion data to identify one or more correlationsbetween the at least a portion of the plurality of datasets with respect to the at least a portion of the plurality of output parameters.
[0095] The system of any preceding clause, wherein the one or more visualizations comprise an indication of the one or more impacts of the at least a portion of the plurality of datasets with respect to the at least a portion of the plurality of output parameters.
[0096] The system of any preceding clause, wherein the processing system is configured to generate the well completion model by: identifying the at least a portion of the plurality of datasets having a degree of variability greater than an additional portion of the plurality of datasets; determining one or more weights associated with the at least a portion of the plurality of datasets affecting with the at least a portion of the plurality of output parameters; and generating the well completion model based on the one or more weights.
[0097] The foregoing description, for purpose of explanation, has been described with reference to specific embodiments. However, the illustrative discussions above are not intended to be exhaustive or to limit the disclosure to the precise forms disclosed. Many modifications and variations are possible in view of the above teachings. Moreover, the order in which the elements of the methods described herein are illustrated and described may be re-arranged, and / or two or more elements may occur simultaneously. The embodiments were chosen and described in order to best explain the principals of the disclosure and its practical applications, to thereby enable others skilled in the art to best utilize the disclosure and various embodiments with various modifications as are suited to the particular use contemplated.
[0098] Finally, the techniques presented and claimed herein are referenced and applied to material objects and concrete examples of a practical nature that demonstrably improve the present technical field and, as such, are not abstract, intangible or purely theoretical. Further, if any claims appended to the end of this specification contain one or more elements designated as “means for [perform]ing [a function]...” or “step for [perform]ing [a function], . it is intended that such elements are to be interpreted under 35 U.S.C. 112(f).However, for any claims containing elements designated in any other manner, it is intended that such elements are not to be interpreted under 35 U.S.C. 112(f).
Claims
CLAIMS1. A method, comprising:receiving well construction process data (122) comprising a plurality of datasets associated with constructing a plurality of wells over a period of time;receiving well completion data (126) associated with the well construction process data, wherein the well completion data comprises a plurality of output parameters associated with constructing the plurality of wells;generating a well completion model (128) based on the well construction process data and the well completion data, wherein the well completion model comprises well insight data associated with one or more impacts of at least a portion of the plurality of datasets with respect to at least a portion of the plurality of output parameters;generating one or more visualizations (134) based on the well insight data; receiving an input (136) at the one or more visualizations; andsending one or more commands (140) to equipment associated with a well construction project in response to receiving the input, wherein the one or more commands are configured to adjust one or more operations of the equipment.
2. The method of claim 1, wherein the well construction process data comprises drilling data, contextual data, workforce data, subsurface data, environmental data, or any combination thereof.
3. The method of claim 1, wherein the well completion data comprises a time duration for performing a well construction task.
4. The method of claim 1, wherein generating the well completion model comprises applying a machine learning algorithm to the well construction process data and the well completion data to identify one or more correlations between the at least a portion of the plurality of datasets with respect to the at least a portion of the plurality of output parameters.
5. The method of claim 1, wherein the one or more visualizations comprise an indication of the one or more impacts of the at least a portion of the plurality of datasets with respect to the at least a portion of the plurality of output parameters.
6. The method of claim 1, wherein the one or more impacts correspond to one or more amounts of time.
7. The method of claim 1, wherein generating the well completion model comprises:identifying the at least a portion of the plurality of datasets having a degree of variability greater than an additional portion of the plurality of datasets;determining one or more weights associated with the at least a portion of the plurality of datasets affecting with the at least a portion of the plurality of output parameters; and generating the well completion model based on the one or more weights.
8. A non-transitory computer-readable medium, comprising computer-executable instructions that, when executed, are configured to cause a processing system to perform the method of claims 1-7.
9. A system comprising:a storage component comprising computer-executable instructions; anda processing system configured to computer-executable instructions to cause a processing system to perform the method of claims 1-7.
