Training an Artificial Intelligence Model to Predict Bit Wear

TECHNICAL FIELD

[0001]This disclosure relates generally to hydrocarbon exploration, drilling, and production, and more particularly, to predicting bit wear.

BACKGROUND

[0002]During drilling, wear components can degrade or fail. For example, drill bits break and otherwise degrade during drilling. When a drill bit wears out or fails, the drill string is removed from the borehole, the drill bit replaced, and then drilling resumes. Drill bit wear is often not immediately observable.

BRIEF DESCRIPTION OF DRAWINGS

[0003]FIG. 1 shows a workflow for training artificial intelligence models to predict bit wear.

[0004]FIG. 2 is a process flow diagram of a process that enables training artificial intelligence models.

[0005]FIG. 3 illustrates hydrocarbon production operations that include both one or more field operations and one or more computational operations, which exchange information and control exploration for the production of hydrocarbons.

[0006]FIG. 4 is a partial schematic perspective view of an example rig system for drilling and producing a well.

[0007]FIG. 5 is a schematic illustration of an example controller (or control system) for that enables a training an artificial intelligence model to predict bit wear.

DETAILED DESCRIPTION

[0008]During oil and gas well drilling, a drill bit is a tool used to cut, crush, shear, or otherwise displace rock of a formation. A drilling program includes various operations that enable hydrocarbon exploration, drilling, and production. In examples, a drilling program is established by evaluating well geometries, casing programs, mud considerations, well control concerns, initial drill bit selections, data from offset wells, pore pressure estimations, economics, and special procedures that may be needed during the life of a well. In examples, bit wear predictions dictate planned drill string removals in the drilling program. The operations that form the drilling program are subject to change if drilling conditions dictate.

[0009]For a target well (e.g., well being planned/drilled), data associated with offset wells can be used to develop a drilling program including the planned operations at the target well. In examples, an offset well is an existing wellbore close (e.g., within a threshold distance) to a target well. The data associated with the offset well provides information for planning or drilling the target well. In examples, numerous offsets are drilled, so a great deal is known about the subsurface geology and pressure regimes as indicated by data associated with the offset wells. Data associated with offset wells is captured by downhole and surface tools during drilling or derived after drilling.

[0010]Artificial intelligence models, such as machine learning models, are used in hydrocarbon exploration, drilling, and production. The models are trained to draw conclusions using a collection of data. The conclusions include solving problems, making predictions, classifying input data, and the like. In examples, artificial intelligence models predict bit wear associated with drilling a target well. For example, bit wear is predicted when planning a drilling program or during actual drilling operations. The artificial intelligence models are trained using data associated with offset wells. However, the formation and drilling operations can vary significantly from well to well, even in the same field. To accurately predict bit wear for a target well, data associated with offset wells most similar to (e.g., relevant to) the target well are used to train the artificial intelligence model.

[0011]Conventional methods include manual bit wear estimation. However, manual estimates of bit wear are highly subjective and convoluted by changes in formation and drilling data. Conventional methods using a physics-based model and supervised machine learning are time consuming, and their accuracy is significantly limited by the labelled data available. Artificial intelligence models trained using the limited labeled data are unable to robustly extract relevant offset wells, especially where the similarity between the offset well and target well is based on patterns or other similarities undetectable by the single comparison. Moreover, conventional approaches do not consider the entire real-time time/depth series. Conventional methods to estimate bit wear are limited to evaluation of offset well data captured at the beginning of a run and the end of the run. In examples, a run refers to a series of events measured or observed during hydrocarbon exploration, drilling, and production. In examples, with reference to logging, a run is an operation in which a logging tool is lowered into a borehole and then retrieved from it while recording measurements. A well can be associated with multiple runs at varying depths.

[0012]The present techniques compares a distribution of data across the entire run, and projects the raw data to a reduced dimension space (called latent space) through the trained AI model. The projection of raw data yields previously hidden patterns or features of the data, which is used to determine a similarity between offset wells and a target well. Accordingly, the present techniques train an artificial intelligence model to predict bit wear using filtered data. The filtered data improves an accuracy of bit wear predictions made using trained artificial intelligence models. In particular, the filters used to extract data associated with relevant offset wells enables a more robust prediction of bit wear when drilling a target well. A precisely filtered dataset associated with offset wells forms the training data used to train the artificial intelligence model. This filtered dataset is developed by applying physics based filters and data driven filters to obtained data associated with an offset well. In examples, the applying physics based filters and data driven filters are applied to the obtained data associated with an offset well in parallel.

[0013]FIG. 1 shows a workflow for training artificial intelligence models to predict bit wear. In examples, a drill bit is located at the bottom end of a drill string, and is changed when it becomes excessively dull or stops making process through the formation. An artificial intelligence model is trained to predict the wear on a drill bit, such as how damaged, worn, or eroded the drill bit has become over time. In examples, drill bits work by scraping or crushing the rock, or both, as a part of a rotational motion. In examples, the drill bits are hammer bits, and pound the rock vertically to displace the rock. Over time, drill bits will become damaged, worn, eroded, or excessively dull. If the bit has enough wear, the bit will stop making process through the formation. An artificial intelligence model is trained to predict bit wear. In some embodiments, the artificial intelligence model is trained using filtered data prior to drilling a target well. For example, the artificial intelligence model is trained to predict bit wear as part of developing a drilling program. In some embodiments, the artificial intelligence model is trained using filtered data while drilling a target well. For example, the artificial intelligence model is trained to predict bit wear during ongoing drilling operations, in real time.

[0014]In examples, the bit wear predictions output by the trained artificial intelligence models are Boolean values that indicate bit wear. In examples, zero (0) represents a drill bit with some remaining useful life, and one (1) represents a dull damaged, or otherwise failed drill bit. In examples, the bit wear predictions output by the trained artificial intelligence models are a probability distribution over a set of classes indicating a level of severity of bit wear. The predicted class is the class with a highest probability. For example, the classes are sharp drill bit, slightly worn drill bit, heavily worn drill bit, and damaged drill bit.

[0015]FIG. 1 shows a data pool 110 associated with offset wells. To train an artificial intelligence model to predict bit wear, data associated with offset wells is obtained. In examples the data associated with offset wells includes drilling parameters, and rock information. For example, the drilling parameters can include weight on bit (WOB), rate of penetration (ROP), rotations per minute (RPM), flow rate, torque and the like. Rock information can include rock unconfined compressive strength (UCS), and rock logging data such as density, porosity, and resistivity loggings. In examples, the offset runs data pool 110 includes a collection of drilling data associated with offset runs. Drilling data can come from many sources, such as sensors, logs, cores, and tests. In examples, sensors measure physical parameters, such as pressure, temperature, torque, and vibration, during drilling. Logs are records of the properties of the formations and fluids encountered by the drill bit at various depths, such as resistivity, porosity, and density. Cores are cylindrical samples of the rock at various depths. The cores are extracted from the borehole for further analysis. Tests are experiments that are performed on the wellbore, the formation, or the fluids to obtain information about their characteristics and behavior.

[0016]In some embodiments, each data associated with a respective offset run includes time-based and depth-based drilling data, log data, dull data, run data (e.g., depth in/out, mud type, bottomhole assembly (BHA)). The offset run data pool 110 is processed before applying the filters. In examples, processing applied to the offset run data pool 110 prior to filtering includes filling in missing data and removing outliers. In examples, the processing includes transforming the offset run data pool 110 to a standard data format. The transformation ensures that data from each run is comparable. For example, the Gamma data from different runs may be measured using different tools with different units. The units are converted to a same unit for runs in the offset data pool 110. Additionally, in examples vertical drilling wells can have varying start and end drilling depths for each runs. Data at the same or similar depths is extracted for comparison.

[0017]Filters are applied to the data associated with offset wells 110. In the example of FIG. 1, two types of filters are shown: physics-based filters 120 and data driven filters 130. For ease of description, two types of filters are described. However, any number of filter types can be used according to the present techniques. Physics based filters 120 can include a comparison between (1) actual data or predicted data associated with the target well from a physics-based model and (2) obtained data associated with an offset well. Data-driven filters 130 are based on data analysis and interpretation of actual data or predicted associated with the target well and data associated with the offset wells. The artificial intelligence based model is trained using filtered data. The reduction of the training set through the filters increases the relevance of the training data with respect to the target well, thereby increasing the accuracy and performance of bit wear predictions associated with the target well. The quality of the bit wear predictions can be further increased by performing an iterative training algorithm, in which the artificial intelligence model is further trained with an updated training set containing data captured during drilling the target well and re-filtered offset training data using the target well data. In examples, when target well data is available, the target well data is compared with data associated with the offset wells. Filtering is then applied to obtain data associated with the offset wells similar to or the same as the target well data. This filtered data enables a robust bit wear prediction model that can predict bit wear accurately.

[0018]In some embodiments, the physics-based filters 120 include a geomechanics model that operates on the data associated with offset wells. In examples, the geomechanics model obtains rock strength and frictional angles from the data associated with offset wells. Additionally, the geomechanics model predicts rock strength and frictional angles for the target well. The offset well data is filtered based on a measure of similarity with the predicted rock strength and frictional angles for the target well. In examples, relevant offset wells have similar rock strength and frictional angles as predicted for the target well. For example, similar rock strength and frictional angles are determined based on a percentage difference between two number values in order to determine how close they are, relative to the target value. In an example, relevant offset wells are within +/−20% of an average predicted rock strength value for the target well. In an example, relevant offset wells are within +/−5 degrees of an average predicted frictional angle for the target well.

[0019]In some embodiments, the physics-based filters 120 includes a lithology model that operates on the data associated with offset wells. In examples, the lithology model obtains lithology and composition from the data associated with offset wells. Additionally, the lithology model predicts lithology and composition for the target well. The offset well data is filtered based on a measure of similarity with the predicted lithology and composition for the target well. In examples, relevant offset wells have similar lithology and composition as predicted for the target well. Similar lithology and composition are determined based on a percentage difference between two number values in order to determine how close they are, relative to the target value. In examples, a comparison is made between the percentage of different types of rock between two runs. Consider an example where offset run A has 90% sandstone, 10% of shale, and offset run B has 90% shale, 10% of sandstone. For a target well that is has a predicted lithology of 80% sandstone and 20% shale, offset run A may be considered a relevant run as it mostly includes sandstone.

[0020]In some embodiments, the physics-based filters 120 include the rate of penetration or a wear baseline model. The wear baseline model that predicts wear on a drill bit when drilling a target well. In examples, the wear on a drill bit when drilling a target well is based on a rate of penetration. The offset well data includes a known rate of penetration and bit wear data, as the offset well data includes known ground truths. In examples, relevant offset wells have similar rate of penetration and bit wear data as predicted for the target well. Similar rate of penetration and bit wear data are determined based on a percentage difference between two number values in order to determine how close they are, relative to the target value. In an example, relevant offset wells are within +/−50% of an average predicted rate of penetration value for the target well. Consider an example where a planned rate of penetration for the target well is 15-20 ft/hr. An offset well with a rate of penetration of 40-60 ft/hr is not a relevant offset well since the rates of penetration are different or outside of a predetermined threshold. Consider an example where a target well is expected to have smooth wear with dull grade 1-4. An offset well with final dull 6-8 is not a relevant offset well since the wear is different.

[0021]In some embodiments, the data driven filters 130 compare a distribution of data across the entire run and project the raw data to a latent space using a trained artificial intelligence model. The latent space has a reduced dimensionality when compared with the dimensionality of the raw data. Reducing the dimensionality of the data yields previously hidden patterns or features of the data that are used to determine a similarity between offset wells and a target well. In examples, the artificial intelligence model is an unsupervised learning model trained using a few offset runs from a predetermined area. The trained model is applied to an entire group of offset runs in the predetermined area. Then among the entire group of offset runs, the data distributions of each run in the latent space are compared. In examples, each run is associated with hundreds or thousands of data points in the projected latent space at different drilling depths. If an offset run has a very different distribution in the latent space when compared with other runs in the group, it can be considered an outlier.

[0022]In some embodiments, the data driven filters 130 include a statistical distribution of logging data from the offset wells and a statistical distribution of logging data predicted for the target well. In examples, the logging data includes gamma, porosity, density, resistivity, and so on. In examples, similar logging data are determined based on a percentage difference between two number values at similar or the same depth across the statistical distribution in order to determine how close they are, relative to the target value. For example, a difference of in distributions between target well and offset well is less than 30%.

[0023]In some embodiments, the data driven filters 130 include a statistical distribution of drilling parameters from the offset wells and a statistical distribution of drilling parameters predicted for the target well. In examples, the drilling parameters include weight on bit (WOB), rate of penetration (ROP), rotations per minute (RPM), and flow rate. In examples, similar drilling parameters are determined based on a percentage difference between the distributions in order to determine how close they are, relative to the target value. For example, a difference in distributions between target well and offset well is less than 30%.

[0024]In some embodiments, the data driven filters 130 include a comparison of well information from the offset wells and well information from the target well. In examples, the well information includes distance to the target, BHA/tools, hole size, and drilling fluid. For example, if the target well uses mud motor, then relevant offset wells should also include mud motor. If target well uses oil based mud, the offset wells with water based mud should be filtered from the training data.

[0025]Applying the physics based filters 120 to the data pool 110 associated with offset runs results in the first filtered dataset 140. Applying the data driven filters 130 to the data pool 110 associated with offset runs results in the second filtered dataset 150. A training dataset 160 of relevant offset wells is created from the first filtered dataset 140 and the second filtered dataset 150. For a first round of training an artificial intelligence model, runs occurring in both of the first filtered dataset 140 and the second filtered dataset 150 are stored in the training dataset 160. In examples, if a particular run is filtered from at least one of the first filtered dataset or the second filtered dataset, the run is not included in the training dataset 160. In some embodiments, a particular run filtered from one of the first filtered dataset or the second filtered dataset is stored in a backup dataset. If the prediction accuracy of a model trained using training dataset 160 the is not satisfactory, the backup dataset can be used to further train the model.

[0026]FIG. 2 is a process flow diagram of a process 200 that enables training artificial intelligence models. The process flow diagram 200 can govern the hydrocarbon production operations 300 of FIG. 3. In some embodiments, the data captured during the hydrocarbon production operations 300 of FIG. 3 is added to a data pool and filtered to train an artificial intelligence model. The process 200 can be implemented by the controller 500 of FIG. 5.

[0027]At block 202, one or more filters is applied to data associated with offset wells. In some embodiments, the data associated with offset wells is preprocessed prior to the application of filters. In example, preprocessing includes standardizing, normalizing, or otherwise formatting the data. In examples, physics-based filters and data driven filters are applied to the data associated with offset wells. For example, a physics-based filter compares the value ranges and correlations of the derived physical properties using physics models, such as a geomechanics model or lithology model. In examples, the physics-based filtering includes executing physics-based simulations, such as bit wear and ROP models, associated with the target well. The simulated target well values are used to filter the data associated with offset wells. In examples, a progressive wear trajectory of different offset runs is compared with the simulated values associated with the target well.

[0028]In some embodiments, the physics based filter includes, for example: 1) a geomechanics model to obtain rock strength/frictional angle and compare between target well and offset wells for filtering; 2) a lithology model to compare rock composition for filtering; 3) a ROP and/or wear baseline model to compare the expected target well performance and offset well for filtering; 4) other physics property comparison using physics-based model; or any combinations thereof.

[0029]In some embodiments, the data-driven filters include comparing value ranges and correlations of the raw surface/downhole drilling and logging while drilling (LWD) measurements of different runs. The data driven filters include, for example: 1) a statistical distribution of offset logging data (LWD/MWD) and expected target well for filtering; 2) a statistical distribution of offset drilling parameters (WOB/RPM/ROP) and expected target well drilling parameters; 3) well information comparison and filtering: distance to the target well; BHA/tools; hole size, drilling fluid; or any combinations thereof.

[0030]At block 204, a first filtered dataset including the set of offset wells filtered by the physics-based filter (e.g., a first set of relevant offset wells) is created. In the first filtered dataset, offset runs that are irrelevant or dissimilar to the target well based on physics based filtering are removed from the dataset.

[0031]At block 206, a second filtered dataset including the set of offset wells filtered by the data driven filter (e.g., a second set of relevant offset wells) is created. In the second filtered dataset, offset runs that are irrelevant or dissimilar to the target well based on data driven filtering are removed from the dataset.

[0032]At block 208, a training set of relevant offset wells is created from the first set of relevant offset wells and the second set of relevant offset wells.

[0033]At block 210, an artificial intelligence model is trained to predict bit wear using the training set of relevant offset wells. In some embodiments, the artificial intelligence model is trained and validated prior to drilling the target well. In some embodiments, the artificial intelligence model is trained in real time while drilling the target well. In examples, the quality of the bit wear predictions can be further increased by performing an iterative training algorithm, in which the artificial intelligence model is further trained with an updated training set containing data captured during drilling the target well. This filtered data enables a robust bit wear prediction model that can predict bit wear accurately.

[0034]In examples, the predictions of bit wear dictate drilling operations. For example, adjustments to drilling parameters such as weight on bit (WOB), rotational speed (RPM), and drilling fluid flow rate are made to compensate for the reduced cutting efficiency of the worn bit according to the predicted bit wear. Alternatively or in addition, in examples, adjustments to planned drilling paths in the drilling program are made based on the predicted bit wear. Further, predictions of bit wear are used to determine planned drill string removals to replace the bit.

[0035]FIG. 3 illustrates hydrocarbon production operations 300 that include both one or more field operations 310 and one or more computational operations 312, which exchange information and control exploration for the production of hydrocarbons. In some implementations, the artificial intelligence model can be trained and/or output generated before, during, or in combination with the hydrocarbon production operations 300, specifically, for example, either as field operations 310 or computational operations 312, or both.

[0036]Examples of field operations 310 include forming/drilling a wellbore, hydraulic fracturing, producing through the wellbore, injecting fluids (such as water) through the wellbore, to name a few. In some implementations, methods of the present disclosure can trigger or control the field operations 310. For example, the methods of the present disclosure can generate data from hardware/software including sensors and physical data gathering equipment (e.g., seismic sensors, well logging tools, flow meters, and temperature and pressure sensors). The methods of the present disclosure can include transmitting the data from the hardware/software to the field operations 310 and responsively triggering the field operations 310 including, for example, generating plans and signals that provide feedback to and control physical components of the field operations 310. Alternatively or in addition, the field operations 310 can trigger the methods of the present disclosure. For example, implementing physical components (including, for example, hardware, such as sensors) deployed in the field operations 310 can generate plans and signals that can be provided as input or feedback (or both) to the methods of the present disclosure.

[0037]Examples of computational operations 312 include one or more computer systems 320 that include one or more processors and computer-readable media (e.g., non-transitory computer-readable media) operatively coupled to the one or more processors to execute computer operations to perform the methods of the present disclosure. The computational operations 312 can be implemented using one or more databases 318, which store data received from the field operations 310 and/or generated internally within the computational operations 312 (e.g., by implementing the methods of the present disclosure) or both. For example, the one or more computer systems 320 process inputs from the field operations 310 to assess conditions in the physical world, the outputs of which are stored in the databases 318. For example, seismic sensors of the field operations 310 can be used to perform a seismic survey to map subterranean features, such as facies and faults. In performing a seismic survey, seismic sources (e.g., seismic vibrators or explosions) generate seismic waves that propagate in the earth and seismic receivers (e.g., geophones) measure reflections generated as the seismic waves interact with boundaries between layers of a subsurface formation. The source and received signals are provided to the computational operations 312 where they are stored in the databases 318 and analyzed by the one or more computer systems 320.

[0038]In some implementations, one or more outputs 322 generated by the one or more computer systems 320 can be provided as feedback/input to the field operations 310 (either as direct input or stored in the databases 318). The field operations 310 can use the feedback/input to control physical components used to perform the field operations 310 in the real world.

[0039]For example, the computational operations 312 can process the seismic data to generate three-dimensional (3D) maps of the subsurface formation. The computational operations 312 can use these 3D maps to provide plans for locating and drilling exploratory wells. In some operations, the exploratory wells are drilled using logging-while-drilling (LWD) techniques which incorporate logging tools into the drill string. LWD techniques can enable the computational operations 312 to process new information about the formation and control the drilling to adjust to the observed conditions in real-time.

[0040]The one or more computer systems 320 can update the 3D maps of the subsurface formation as information from one exploration well is received and the computational operations 312 can adjust the location of the next exploration well based on the updated 3D maps. Similarly, the data received from production operations can be used by the computational operations 312 to control components of the production operations. For example, production well and pipeline data can be analyzed to predict slugging in pipelines leading to a refinery and the computational operations 312 can control machine operated valves upstream of the refinery to reduce the likelihood of plant disruptions that run the risk of taking the plant offline.

[0041]In some implementations of the computational operations 312, customized user interfaces can present intermediate or final results of the above-described processes to a user. Information can be presented in one or more textual, tabular, or graphical formats, such as through a dashboard. The information can be presented at one or more on-site locations (such as at an oil well or other facility), on the Internet (such as on a webpage), on a mobile application (or app), or at a central processing facility.

[0042]The presented information can include feedback, such as changes in parameters or processing inputs, that the user can select to improve a production environment, such as in the exploration, production, and/or testing of petrochemical processes or facilities. For example, the feedback can include parameters that, when selected by the user, can cause a change to, or an improvement in, drilling parameters (including drill bit speed and direction) or overall production of a gas or oil well. The feedback, when implemented by the user, can improve the speed and accuracy of calculations, streamline processes, improve models, and solve problems related to efficiency, performance, safety, reliability, costs, downtime, and the need for human interaction.

[0043]In some implementations, the feedback can be implemented in real-time, such as to provide an immediate or near-immediate change in operations or in a model. The term real-time (or similar terms as understood by one of ordinary skill in the art) means that an action and a response are temporally proximate such that an individual perceives the action and the response occurring substantially simultaneously. For example, the time difference for a response to display (or for an initiation of a display) of data following the individual's action to access the data can be less than 1 millisecond (ms), less than 1 second(s), or less than 5 s. While the requested data need not be displayed (or initiated for display) instantaneously, it is displayed (or initiated for display) without any intentional delay, taking into account processing limitations of a described computing system and time required to, for example, gather, accurately measure, analyze, process, store, or transmit the data.

[0044]Events can include readings or measurements captured by downhole equipment such as sensors, pumps, bottom hole assemblies, or other equipment. The readings or measurements can be analyzed at the surface, such as by using applications that can include modeling applications and machine learning. The analysis can be used to generate changes to settings of downhole equipment, such as drilling equipment. In some implementations, values of parameters or other variables that are determined can be used automatically (such as through using rules) to implement changes in oil or gas well exploration, production/drilling, or testing. For example, outputs of the present disclosure can be used as inputs to other equipment and/or systems at a facility. This can be especially useful for systems or various pieces of equipment that are located several meters or several miles apart, or are located in different countries or other jurisdictions.

[0045]In some embodiments, the one or more field operations 310 and one or more computational operations 312 controls drilling tools and equipment for performing drilling actions in a wellbore. Some components of an example drilling system are shown in FIG. 4. Among other components, the drilling system 312 can include a drilling rig, a drill string, and a bottomhole assembly (BHA). The BHA includes tools for tools that are attached to the end of the drill string that are used for drilling, coring, and logging operations.

[0046]FIG. 4 is a partial schematic perspective view of an example rig system 400 for drilling and producing a well. The well is drilled using a drill bit, and bit wear is predicted according to the present techniques. The well can extend from the surface through the Earth to one or more subterranean zones of interest. The example rig system 400 includes a drill floor 402 positioned above the surface, a wellhead 404, a drill string assembly 406 supported by the rig structure, a fluid circulation system 408 to filter used drilling fluid from the wellbore and provide clean drilling fluid to the drill string assembly 406. For example, the example rig system 400 of FIG. 4 is shown as a drill rig capable of performing a drilling operation with the rig system 400 supporting the drill string assembly 406 over a wellbore. The wellhead 404 can be used to support casing or other well components or equipment into the wellbore of the well.

[0047]The derrick or mast is a support framework mounted on the drill floor 402 and positioned over the wellbore to support the components of the drill string assembly 406 during drilling operations. A crown block 412 forms a longitudinally-fixed top of the derrick, and connects to a travelling block 414 with a drilling line including a set of wire ropes or cables. The crown block 412 and the travelling block 414 support the drill string assembly 406 via a swivel 416, a kelly 418, or a top drive system (not shown). Longitudinal movement of the travelling block 414 relative to the crown block 412 of the drill string assembly 406 acts to move the drill string assembly 406 longitudinally upward and downward. The swivel 416, connected to and hung by the travelling block 414 and a rotary hook, allows free rotation of the drill string assembly 406 and provides a connection to a kelly hose 420, which is a hose that flows drilling fluid from a drilling fluid supply of the circulation system 408 to the drill string assembly 406. A standpipe 422 mounted on the drill floor 402 guides at least a portion of the kelly hose 420 to a location proximate to the drill string assembly 406. The kelly 418 is a hexagonal device suspended from the swivel 416 and connected to a longitudinal top of the drill string assembly 406, and the kelly 418 turns with the drill string assembly 406 as the rotary table 442 of the drill string assembly turns.

[0048]In the example rig system 400 of FIG. 4, the drill string assembly 406 is made up of drill pipes with a drill bit (not shown) at a longitudinally bottom end of the drill string. The drill pipe can include hollow steel piping, and the drill bit can include cutting tools, such as blades, discs, rollers, cutters, or a combination of these, to cut into the formation and form the wellbore. The drill bit rotates and penetrates through rock formations below the surface under the combined effect of axial load and rotation of the drill string assembly 406. In some implementations, the kelly 418 and swivel 416 can be replaced by a top drive that allows the drill string assembly 406 to spin and drill. The wellhead assembly 404 can also include a drawworks 424 and a deadline anchor 426, where the drawworks 424 includes a winch that acts as a hoisting system to reel the drilling line in and out to raise and lower the drill string assembly 406 by a fast line 425. The deadline anchor 426 fixes the drilling line opposite the drawworks 424 by a deadline 427, and can measure the suspended load (or hook load) on the rotary hook. The weight on bit (WOB) can be measured when the drill bit is at the bottom the wellbore. The wellhead assembly 404 also includes a blowout preventer 450 positioned at the surface of the well and below (but often connected to) the drill floor 402. The blowout preventer 450 acts to prevent well blowouts caused by formation fluid entering the wellbore, displacing drilling fluid, and flowing to the surface at a pressure greater than atmospheric pressure. The blowout preventer 450 can close around (and in some instances, through) the drill string assembly 406 and seal off the space between the drill string and the wellbore wall.

[0049]During a drilling operation of the well, the circulation system 408 circulates drilling fluid from the wellbore to the drill string assembly 406, filters used drilling fluid from the wellbore, and provides clean drilling fluid to the drill string assembly 406. The example circulation system 408 includes a fluid pump 430 that fluidly connects to and provides drilling fluid to drill string assembly 406 via the kelly hose 420 and the standpipe 422. The circulation system 408 also includes a flow-out line 432, a shale shaker 434, a settling pit 436, and a suction pit 438. In a drilling operation, the circulation system 408 pumps drilling fluid from the surface, through the drill string assembly 406, out the drill bit and back up the annulus of the wellbore, where the annulus is the space between the drill pipe and the formation or casing. The density of the drilling fluid is intended to be greater than the formation pressures to prevent formation fluids from entering the annulus and flowing to the surface and less than the mechanical strength of the formation, as a greater density may fracture the formation, thereby creating a path for the drilling fluids to go into the formation. Apart from well control, drilling fluids can also cool the drill bit and lift rock cuttings from the drilled formation up the annulus and to the surface to be filtered out and treated before it is pumped down the drill string assembly 406 again. The drilling fluid returns in the annulus with rock cuttings and flows out to the flow-out line 432, which connects to and provides the fluid to the shale shaker 434. The flow line is an inclined pipe that directs the drilling fluid from the annulus to the shale shaker 434. The shale shaker 434 includes a mesh-like surface to separate the coarse rock cuttings from the drilling fluid, and finer rock cuttings and drilling fluid then go through the settling pit 436 to the suction pit 436. The circulation system 408 includes a mud hopper 440 into which materials (for example, to provide dispersion, rapid hydration, and uniform mixing) can be introduced to the circulation system 408. The fluid pump 430 cycles the drilling fluid up the standpipe 422 through the swivel 416 and back into the drill string assembly 406 to go back into the well.

[0050]The example wellhead assembly 404 can take a variety of forms and include a number of different components. For example, the wellhead assembly 404 can include additional or different components than the example shown in FIG. 4. Similarly, the circulation system 408 can include additional or different components than the example shown in FIG. 4.

[0051]FIG. 5 is a schematic illustration of an example controller 500 (or control system) for that enables training an artificial intelligence model to predict bit wear. For example, the controller 500 may be operable according to the workflow 100 of FIG. 1 or the process 200 of FIG. 2. In some embodiments, the controller 500 is the same as or similar to the computer systems 320 of FIG. 3. The controller 500 is intended to include various forms of digital computers, such as printed circuit boards (PCB), processors, digital circuitry, or otherwise parts of a system for supply chain alert management. Additionally the system can include portable storage media, such as, Universal Serial Bus (USB) flash drives. For example, the USB flash drives may store operating systems and other applications. The USB flash drives can include input/output components, such as a wireless transmitter or USB connector that may be inserted into a USB port of another computing device.

[0052]The controller 500 includes a processor 510, a memory 520, a storage device 530, and an input/output interface 540 communicatively coupled with input/output devices 560 (for example, displays, keyboards, measurement devices, sensors, valves, pumps). Each of the components 510, 520, 530, and 540 are interconnected using a system bus 550. The processor 510 is capable of processing instructions for execution within the controller 500. The processor may be designed using any of a number of architectures. For example, the processor 510 may be a CISC (Complex Instruction Set Computers) processor, a RISC (Reduced Instruction Set Computer) processor, or a MISC (Minimal Instruction Set Computer) processor.

[0053]In one implementation, the processor 510 is a single-threaded processor. In another implementation, the processor 510 is a multi-threaded processor. The processor 510 is capable of processing instructions stored in the memory 520 or on the storage device 530 to display graphical information for a user interface on the input/output interface 540.

[0054]The memory 520 stores information within the controller 500. In one implementation, the memory 520 is a computer-readable medium. In one implementation, the memory 520 is a volatile memory unit. In another implementation, the memory 520 is a nonvolatile memory unit.

[0055]The storage device 530 is capable of providing mass storage for the controller 500. In one implementation, the storage device 530 is a computer-readable medium. In various different implementations, the storage device 530 may be a floppy disk device, a hard disk device, an optical disk device, or a tape device.

[0056]The input/output interface 540 provides input/output operations for the controller 500. In one implementation, the input/output devices 560 includes a keyboard and/or pointing device. In another implementation, the input/output devices 560 includes a display unit for displaying graphical user interfaces.

[0057]There can be any number of controllers 500 associated with, or external to, a computer system containing controller 500, with each controller 500 communicating over a network. Further, the terms “client,” “user,” and other appropriate terminology can be used interchangeably, as appropriate, without departing from the scope of the present disclosure. Moreover, the present disclosure contemplates that many users can use one controller 500 and one user can use multiple controllers 500.

Embodiments

[0058]According to some non-limiting embodiments or examples, provided is a computer-implemented method that enables training an artificial intelligence model to predict bit wear, including: applying multiple filters to a data pool associated with offset wells, where the filters include physics-based filters and data driven filters; creating a first filtered dataset including data associated with a set of offset wells filtered by the physics-based filter; creating a second filtered dataset including data associated with a set of offset wells filtered by the data driven filter; creating a training dataset of relevant offset wells from the first filtered dataset and the second filtered dataset; and training an artificial intelligence model to predict bit wear at a target well using the training dataset of relevant offset wells.

[0059]According to some non-limiting embodiments or examples, provided is an apparatus including a non-transitory, computer readable, storage medium that stores instructions that, when executed by at least one processor, cause the at least one processor to perform operations including: applying multiple filters to a data pool associated with offset wells, where the filters include physics-based filters and data driven filters; creating a first filtered dataset including data associated with a set of offset wells filtered by the physics-based filter; creating a second filtered dataset including data associated with a set of offset wells filtered by the data driven filter; creating a training dataset of relevant offset wells from the first filtered dataset and the second filtered dataset; and training an artificial intelligence model to predict bit wear at a target well using the training dataset of relevant offset wells.

[0060]According to some non-limiting embodiments or examples, provided is a system, including: one or more memory modules; one or more hardware processors communicably coupled to the one or more memory modules, the one or more hardware processors configured to execute instructions stored on the one or more memory modules to perform operations including: applying multiple filters to a data pool associated with offset wells, where the filters include physics-based filters and data driven filters; creating a first filtered dataset including data associated with a set of offset wells filtered by the physics-based filter; creating a second filtered dataset including data associated with a set of offset wells filtered by the data driven filter; creating a training dataset of relevant offset wells from the first filtered dataset and the second filtered dataset; and training an artificial intelligence model to predict bit wear at a target well using the training dataset of relevant offset wells.

[0082]Implementations of the subject matter and the functional operations described in this specification can be implemented in digital electronic circuitry, in tangibly embodied computer software or firmware, in computer hardware, including the structures disclosed in this specification and their structural equivalents, or in combinations of one or more of them. Software implementations of the described subject matter can be implemented as one or more computer programs. Each computer program can include one or more modules of computer program instructions encoded on a tangible, non-transitory, computer-readable computer-storage medium for execution by, or to control the operation of, data processing apparatus. Alternatively, or additionally, the program instructions can be encoded in/on an artificially generated propagated signal. The example, the signal can be a machine-generated electrical, optical, or electromagnetic signal that is generated to encode information for transmission to suitable receiver apparatus for execution by a data processing apparatus. The computer-storage medium can be a machine-readable storage device, a machine-readable storage substrate, a random or serial access memory device, or a combination of computer-storage mediums.

[0083]The terms “data processing apparatus,” “computer,” and “electronic computer device” (or equivalent as understood by one of ordinary skill in the art) refer to data processing hardware. For example, a data processing apparatus can encompass all kinds of apparatus, devices, and machines for processing data, including by way of example, a programmable processor, a computer, or multiple processors or computers. The apparatus can also include special purpose logic circuitry including, for example, a central processing unit (CPU), a field programmable gate array (FPGA), or an application specific integrated circuit (ASIC). In some implementations, the data processing apparatus or special purpose logic circuitry (or a combination of the data processing apparatus or special purpose logic circuitry) can be hardware- or software-based (or a combination of both hardware- and software-based). The apparatus can optionally include code that creates an execution environment for computer programs, for example, code that constitutes processor firmware, a protocol stack, a database management system, an operating system, or a combination of execution environments. The present disclosure contemplates the use of data processing apparatuses with or without conventional operating systems, for example, LINUX, UNIX, WINDOWS, MAC OS, ANDROID, or IOS.

[0084]A computer program, which can also be referred to or described as a program, software, a software application, a module, a software module, a script, or code, can be written in any form of programming language. Programming languages can include, for example, compiled languages, interpreted languages, declarative languages, or procedural languages. Programs can be deployed in any form, including as stand-alone programs, modules, components, subroutines, or units for use in a computing environment. A computer program can, but need not, correspond to a file in a file system. A program can be stored in a portion of a file that holds other programs or data, for example, one or more scripts stored in a markup language document, in a single file dedicated to the program in question, or in multiple coordinated files storing one or more modules, sub programs, or portions of code. A computer program can be deployed for execution on one computer or on multiple computers that are located, for example, at one site or distributed across multiple sites that are interconnected by a communication network. While portions of the programs illustrated in the various figures may be shown as individual modules that implement the various features and functionality through various objects, methods, or processes, the programs can instead include a number of sub-modules, third-party services, components, and libraries. Conversely, the features and functionality of various components can be combined into single components as appropriate. Thresholds used to make computational determinations can be statically, dynamically, or both statically and dynamically determined.

[0085]The methods, processes, or logic flows described in this specification can be performed by one or more programmable computers executing one or more computer programs to perform functions by operating on input data and generating output. The methods, processes, or logic flows can also be performed by, and apparatus can also be implemented as, special purpose logic circuitry, for example, a CPU, an FPGA, or an ASIC.

[0086]Computers suitable for the execution of a computer program can be based on one or more of general and special purpose microprocessors and other kinds of CPUs. The elements of a computer are a CPU for performing or executing instructions and one or more memory devices for storing instructions and data. Generally, a CPU can receive instructions and data from (and write data to) a memory. A computer can also include, or be operatively coupled to, one or more mass storage devices for storing data. In some implementations, a computer can receive data from, and transfer data to, the mass storage devices including, for example, magnetic, magneto optical disks, or optical disks. Moreover, a computer can be embedded in another device, for example, a mobile telephone, a personal digital assistant (PDA), a mobile audio or video player, a game console, a global positioning system (GPS) receiver, or a portable storage device such as a universal serial bus (USB) flash drive.

[0087]Computer readable media (transitory or non-transitory, as appropriate) suitable for storing computer program instructions and data can include all forms of permanent/non-permanent and volatile/non-volatile memory, media, and memory devices. Computer readable media can include, for example, semiconductor memory devices such as random access memory (RAM), read only memory (ROM), phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), and flash memory devices. Computer readable media can also include, for example, magnetic devices such as tape, cartridges, cassettes, and internal/removable disks. Computer readable media can also include magneto optical disks and optical memory devices and technologies including, for example, digital video disc (DVD), CD ROM, DVD+/−R, DVD-RAM, DVD-ROM, HD-DVD, and BLURAY. The memory can store various objects or data, including caches, classes, frameworks, applications, modules, backup data, jobs, web pages, web page templates, data structures, database tables, repositories, and dynamic information. Types of objects and data stored in memory can include parameters, variables, algorithms, instructions, rules, constraints, and references. Additionally, the memory can include logs, policies, security or access data, and reporting files. The processor and the memory can be supplemented by, or incorporated in, special purpose logic circuitry.

[0088]Implementations of the subject matter described in the present disclosure can be implemented on a computer having a display device for providing interaction with a user, including displaying information to (and receiving input from) the user. Types of display devices can include, for example, a cathode ray tube (CRT), a liquid crystal display (LCD), a light-emitting diode (LED), and a plasma monitor. Display devices can include a keyboard and pointing devices including, for example, a mouse, a trackball, or a trackpad. User input can also be provided to the computer through the use of a touchscreen, such as a tablet computer surface with pressure sensitivity or a multi-touch screen using capacitive or electric sensing. Other kinds of devices can be used to provide for interaction with a user, including to receive user feedback including, for example, sensory feedback including visual feedback, auditory feedback, or tactile feedback. Input from the user can be received in the form of acoustic, speech, or tactile input. In addition, a computer can interact with a user by sending documents to, and receiving documents from, a device that is used by the user. For example, the computer can send web pages to a web browser on a user's client device in response to requests received from the web browser.

[0089]While this specification contains many specific implementation details, these should not be construed as limitations on the scope of what may be claimed, but rather as descriptions of features that may be specific to particular implementations. Certain features that are described in this specification in the context of separate implementations can also be implemented, in combination, in a single implementation. Conversely, various features that are described in the context of a single implementation can also be implemented in multiple implementations, separately, or in any suitable sub-combination. Moreover, although previously described features may be described as acting in certain combinations and even initially claimed as such, one or more features from a claimed combination can, in some cases, be excised from the combination, and the claimed combination may be directed to a sub-combination or variation of a sub-combination.

[0090]Particular implementations of the subject matter have been described. Other implementations, alterations, and permutations of the described implementations are within the scope of the following claims as will be apparent to those skilled in the art. While operations are depicted in the drawings or claims in a particular order, this should not be understood as requiring that such operations be performed in the particular order shown or in sequential order, or that all illustrated operations be performed (some operations may be considered optional), to achieve desirable results. In certain circumstances, multitasking or parallel processing (or a combination of multitasking and parallel processing) may be advantageous and performed as deemed appropriate.