SYSTEM AND METHOD OF GENERATING AND IMPLEMENTING OUTPUTS FOR A DRILLING RIG USING A WELL CONSTRUCTION KNOWLEDGE MINING SYSTEM BASED ON WELL CONSTRUCTION DATA

Description

BACKGROUND

Conventionally, during well construction operations, activities on a rig are often dependent on the knowledge of the rig personnel. Decisions based on past experiences of the drilling crew may not be optimal, and can cause rig delays and other impacts to operations.

Knowledge sharing among rig personnel is incomplete in part because, while the data relating to rig operations is often captured in records, the records are not easily identifiable because the records have different formats and are stored in a variety of locations.

Without a knowledge base that is readily available and accessible, it is not feasible to identify meaningful risk factors or other signals associated with the data. Moreover, conventional systems are incapable of generating an output in real-time or near real-time in response to receipt of well construction data and based on the knowledge base.

A drilling rig generates a huge amount of data from various sources, such as sensors, logs, reports, and images. However, most of this data is unstructured, noisy, and incomplete, making it difficult to analyze and interpret. Knowledge mining can help overcome these challenges by applying artificial intelligence (AI) and machine learning (ML) algorithms to transform raw data into meaningful and actionable knowledge.

Knowledge mining is the process of extracting valuable information from large and complex data sets using AI and ML techniques. Knowledge mining can help organizations discover hidden patterns, trends, and insights that can enhance decision making, innovation, and efficiency.

Some of the benefits of knowledge mining on a drilling rig are improved performance, enhanced safety and accelerated technological innovation. Knowledge mining can help optimize drilling parameters, such as speed, pressure, and direction etc. to maximize productivity and minimize costs. Knowledge mining can also help detect and prevent anomalies, such as equipment failures, leaks, and blowouts, to reduce downtime and risks. Knowledge mining can help monitor and control the health and safety of the workers and the environment on the drilling rig. Knowledge mining can also help identify and mitigate hazards, such as fires, explosions, and spills, to protect lives and assets. Knowledge mining can help generate new insights and solutions for complex and novel problems on the drilling rig. Knowledge mining can also help foster a culture of learning and collaboration among the workers and the stakeholders on the drilling rig.

BRIEF DESCRIPTION OF THE DRAWINGS

The present disclosure is best understood from the following detailed description when read with the accompanying figures. It is emphasized that, in accordance with the standard practice in the industry, various features are not drawn to scale. In fact, the dimensions of the various features may be arbitrarily increased or reduced for clarity of discussion.

FIG. 1 is a schematic diagram of a drilling rig apparatus that comprises a computing system in communication with a model via a network, according to one or more aspects of the present disclosure.

FIG. 2 is a data-flow diagram associated with the model of FIG. 1, according to one or more aspects of the present disclosure.

FIG. 3 is a flow-chart diagram of a method according to one or more aspects of the present disclosure.

FIG. 4 is a listing of data used by the model during the method of FIG. 3, according to one or more aspects of the present disclosure.

FIG. 5 is a diagrammatic illustration of portions of the system of FIG. 1 according to one or more aspects of the present disclosure.

FIG. 6 is another flow-chart diagram of a method according to one or more aspects of the present disclosure.

FIG. 7 is a diagrammatic illustration of a node for implementing one or more example embodiments of the present disclosure, according to an example embodiment.

DETAILED DESCRIPTION

It is to be understood that the present disclosure provides many different embodiments, or examples, for implementing different features of various embodiments. Specific examples of components and arrangements are described below to simplify the present disclosure. These are, of course, merely examples and are not intended to be limiting. In addition, the present disclosure may repeat reference numerals and/or letters in the various examples. This repetition is for the purpose of simplicity and clarity and does not in itself dictate a relationship between the various embodiments and/or configurations discussed. Moreover, the formation of a first feature over or on a second feature in the description that follows may include embodiments in which the first and second features are formed in direct contact and may also include embodiments in which additional features may be formed interposing the first and second features, such that the first and second features may not be in direct contact.

Referring to FIG. 1, illustrated is a schematic view of a system 50 that includes an apparatus 100 that is operably coupled to a trained model 105 via a network 110. While FIG. 1 may illustrate the model 105 being remote from the apparatus, in other embodiments the apparatus 100 is operably coupled to the trained model 105 that is executed by a computer system 100l and/or Edge infrastructure, both of which is located at the rig site. In some embodiments, the Edge infrastructure is a highly specialized computing system designed to processes and analyze data and execute multiple computer applications related to drilling operations. Generally, the system 50 provides objective, easy to use, and customizable outputs for use before, during, or after drilling and/or completion operations, which includes providing advisories or recommendations to a user or operator of the apparatus 100 and/or system 50. For example, in some embodiments the trained model 105 provides safety advisories to a crew of the apparatus 100 in an effort to improve safety and improve operational workflow. However, in other embodiments the trained model 105 provides an output upon which instructions are generated and executed by the apparatus 100. Some example objectives include using the system 50 as a safety advisory tool; to drive safety audits, conversations, and videos; and to drive behavioral changes.

In some embodiments, the model 105 includes a machine learning, large language model, or deep learning module that integrates data from a plurality of records and sources, and uses that integrated data to develop and train a machine learning model. In some embodiments, the model 105 includes a deep learning architecture such as deep neural networks. In some embodiment, the model 105 includes a random forest machine learning algorithm. In some embodiments, the machine learning model is or includes neural networks. In some embodiments, the model 105 includes and/or executes one or more web-based programs, Intranet-based programs, and/or any combination thereof. In an example embodiment, the model 105 includes a computer program including a plurality of instructions, data, and/or any combination thereof. In an example embodiment, the computer program is written in, for example, Python, C, or C++. In other embodiments, the computer program may also be written in Hypertext Markup Language (HTML), Cascading Style Sheets (CSS), JavaScript, Extensible Markup Language (XML), asynchronous JavaScript and XML (Ajax), iOS, XCode, Swift, Android for mobile, and/or any combination thereof.

FIG. 2 illustrates a data flow 200 associated with the system 50. As illustrated, data 205 is processed via a data processor 210, and then the model 105, using the data that has been processed, provides an output 215. Generally, the data 205 is well construction data in that the data relates to the construction of a well, whether that is drilling operations, completion operations, etc. Generally, the data 205 includes training data 220, validation data 225, and input data 230. The type of data being processed via the data processor 210 depends on whether the model 105 is in training mode or prediction mode. For example, when the model 105 is in training mode, the training data 220 is processed via the processor 210. Validation data 225 is processed via the data processor 210 before the model 105 is validated. However, input data 230 is processed via the data processor 210 when the model 105 is in prediction mode. In some embodiments, the output 215 includes an advisory, a recommendation, guidance, or an output upon which an advisory, a recommendation, or guidance is based. Generally, the model 105 is trained using historical details of specific events to predict and/or prevent future events from occurring, specifically when the event is a safety event.

Example Training of the Model

In some embodiments, and as illustrated in the flow diagram of FIG. 3, a method 300 of training the model 105 comprises accessing training data at step 305; pre-processing and processing data at step 310; and training and validating the model at step 315.

In some embodiments, the step 305 includes accessing the training data 220, which is a portion of the data 205. Generally, the training data 220 is historical data, such as for example the past four years of data, but other time periods may be included. FIG. 4 is a diagrammatical illustration of the data 205. As illustrated, the data 205 may include equipment failure/maintenance data 205a; alarm data 205b; recipe and well plan data 205c; rig inventory data 205d; safety/HSE data 205e; digital twin data 205f; surface data 205g; downhole data 205h; human resources data 205i; and operations data 205j. Other types of data that is not listed here may also be included in the data 205.

In some embodiments, the equipment failure/maintenance data 205a includes data detailing the type and frequency of maintenance work and the type and frequency of equipment failures. In some embodiments, the equipment failure/maintenance data 205a includes data detailing the types and frequency of routine maintenance checks or tasks recommended and/or preformed on equipment, data gathered during maintenance checks such as for example fluid levels, data gathered during equipment failure reports and replacement requests, data gathered during inspections, etc. The equipment failure/maintenance data 205a can include data for a wide variety of equipment, including hoisting equipment, drill rods, mud pumps, drill bits, engines, hydraulics, filters, hoses, etc. In some embodiments, the equipment failure/maintenance data 205a is generated or captured in a variety of formats, which may include file extensions such as .pdf, .msg, .pst, .docx, etc. The equipment failure/maintenance data 205a may be stored in a variety of physical and virtual locations, depending on the original file format used when the data was created/captured. For example, the equipment failure/maintenance data 205 a may include an image, log files, sensor information etc.

In some embodiments, the alarm data 205b includes data relating to alarms. In some embodiments, the alarm or alert may be a related to a condition of a piece of equipment, may relate to the location of a piece of equipment, may relate to an operator’s location relative to a piece of equipment, may relate to conditions in or around the drilling rig, etc. Similar to the equipment failure/maintenance data 205a, the alarm data 205b can relate to a wide variety of equipment, including hoisting equipment, mud pumps, engines, hydraulics, information, etc. In some embodiments, the alarm data 205b is generated or captured in a variety of formats, which may include file extensions such as .pdf, .msg, .pst, .docx, etc. The alarm data 205b may be stored in a variety of physical and virtual locations, depending on the original file format used when the data was created/captured. For example, computer applications may be used to monitor conditions and create alarms.

In some embodiments, the recipe and the well plan data 205c includes a variety of data relating to a planned wellbore. In some embodiments, the recipe/well plan data 205c includes a recommended surface location or kickoff point for a well, target survey locations for the bottom home assembly (“BHA”) during a drilling operation so that a target well path is created, target drilling parameters during drilling operations, etc. Generally, well plans are designed based on surveys and designed around drilling constraints associated with equipment and/or client preferences. The recipe/well plan data 205c may include target equipment and parameters for specific sections of the wellbore and/or for specific geological formations expected to encounter during the drilling operations. The recipe/well plan data 205c generally not only includes the target plan but also instructions on how to obtains the targets. In some embodiments, the recipe/well plan data 205c is generated or captured in a variety of formats, which may include file extensions such as .pdf, .msg, .pst, .docx, etc. The recipe/well plan data 205c may be stored in a variety of physical and virtual locations, depending on the original file format used when the data was created/captured.

In some embodiments, the rig inventory data 205d includes data detailing the inventory associated with a rig and/or inventories associated with different rigs. The rig inventory includes data relating to the pieces of equipment associated, accessible, or used with the apparatus 100 or rig. In some embodiments, the rig inventory data 205d included data such as for example, the type of drive system, data relating to the draw works, hoist type (e.g., air or hydraulic) near drilling or opposite driller, type, size, and make of mud pumps and tanks, type of driller control panel, type of rig instrumentation and recording system, type/make of catwalk/pipe handler, the type of the rig itself, the substructure details relating to the rig, the mast manufacturer, type, height, and gross normal capacity, winch data, drill line size and type, block make and type, accumulator make, model, and type, valve details such as kill/choke valve type, make, and size, drill pipe type, weight, and quantity, drill collar details, iron roughneck make and type, power tong type, model, make and capacity, drillstring handline equipment, tubing/rod handling equipment, type of generators, fuel and water storage, air compressor details, etc. In some embodiments, the rig inventory data 205d is generated or captured in a variety of formats, which may include file extensions such as .pdf, .msg, .pst, .docx, etc. The rig inventory data 205d may be stored in a variety of physical and virtual locations, depending on the original file format used when the data was created/captured.

In some embodiments, the safety/HSE data 205e includes data relating to an incident, such as incident location, mechanism, and date and time. In some embodiments, the safety/HSE data 205e includes data relating to safety plans, HSE (health, safety, and environment) specific procedures, safety management protocols, equipment compliance, rig floor testing, and pressure testing. The safety/HSE data 205e may include the recommended safety protocols, testing, and procedures, and also data relating to the frequency and feedback associated with completed procedures and tests. In some embodiments, the safety/HSE data 205e includes data obtained via an incident report or the like, such as for example, a description of the incident, the timing and location of the incident, the equipment associated with the incident, the personnel involved in the incident and the personnel that serviced/inspected the equipment associated with the incident (if relevant), the outcome of the incident, etc. In some embodiments, the safety/HSE data 205e is generated or captured in a variety of formats, which may include file extensions such as .pdf, .msg, .pst, .docx, etc. The safety/HSE data 205e may be stored in a variety of physical and virtual locations, depending on the original file format used when the data was created/captured.

In some embodiments, the digital twin data 205f includes data associated with a digital twin as described in U.S. Patent Application No. 18/066,683, the entirety of which is incorporated by reference herein. In some embodiments, the digital twin data 205f includes the data inputs of the digital twin and data outputs of the digital twin.

In some embodiments, the surface data 205g includes data associated with operations, equipment, conditions, and activities at the surface of a well. In some embodiments, the surface data 205g overlaps with the other types of data, such as for example the rig inventory data 205d, equipment failure/maintenance data 205a, etc. In some embodiments, the surface data 205g includes data obtained via cameras located at various positions around the apparatus 100.

In some embodiments, the downhole data 205h includes data associated with operations, equipment, parameters, conditions, and activities within the wellbore. In some embodiments, the downhole data 205h overlaps with the other types of data, such as for example the rig inventory data 205d, equipment failure/maintenance data 205a, etc.

In some embodiments, the human resources data 205i includes crew or staff information, such as an amount of time at a position and job title. In some embodiments, the human resources data 205i includes the work schedule associated with one or more crew or staff of one or more of the rigs. As such, and in some embodiments, the human resources data 205i includes data relating to the number of hours on each shift, the number of days worked consecutively, etc. In some embodiments, the human resources data 205i is generated or captured in a variety of formats, which may include file extensions such as .pdf, .msg, .pst, .docx, etc. The human resources data 205i may be stored in a variety of physical and virtual locations, depending on the original file format used when the data was created/captured.

In some embodiments, the operations data 205j includes data relating to the operations at the rig, which may include the drilling parameters for drilling operations, etc. In some embodiments, the operations data 205j is generated or captured in a variety of formats, which may include file extensions such as .pdf, .msg, .pst, .docx, etc. The operations data 205j may be stored in a variety of physical and virtual locations, depending on the original file format used when the data was created/captured.

An example record that is used to train the model 105 is a historical drilling record, which may include the data 205a-205j referenced above. In some embodiments, the historical drilling record may include surface data monitored or tracked during the drilling of a wellbore; client name data associated with the wellbore; formation data (expected and encountered) associated with the wellbore; drilling equipment data associated with the drilling of the wellbore; wellbore segment(s) data; field or geographical area data associated with the field or geographical area in which the wellbore was drilled; shape of the wellbore data (i.e., straight or vertical, L-shape, S-shape, etc.); target drilling parameters data associated with the target drilling parameters used during the drilling of the wellbore; geographical limitations data associated with the drilling of the wellbore; survey details or downhole data obtained during the drilling of the wellbore; engineering data; and other data.

With respect to the step 305 of the method 300, in some embodiments, the training data 220 is accessed by the model 105. In some embodiments, and as noted above, the training data 220 data may be stored in a variety of locations and in a variety of formats. In some embodiments, the data or a portion thereof is unstructured data. The location and/or format of the data often depends on the type of data. For example, downhole data is generally initially stored in a tool that is placed in a wellbore that extends from a drilling rig, then, upon tripping out of the tool or upon the data being otherwise transferred to the surface (via electronic transmission through a wireline or wired pipe, and/or transmission as electromagnetic pulses for example) may be transferred to a location that is near or at a drilling rig, then copied or transferred to an offsite (relative to the drilling rig) server. In this example, the format of the data may change. For example, the data format when stored in the tool may be different from the data format when stored at the surface, etc. Another example relates to alarm data 205b, which may include data relating to the values that were being monitored that triggered an alarm, data relating to the generation of the alarm itself (e.g., was an alarm generated, what kind of alarm was generated, when was the alarm generated), and data relating to the reaction of the alarm (e.g., was the alarm acknowledged, were conditions altered to affect/correct the values that originally triggered the alarm). Another example relates to recipe/well plan data 205c, which may be in the form of an email, pdf, and/or electronic application. In some embodiments, the model 105 accesses the training data 220 via an interface specific to each format and/or location associated with the training data 220.

In some embodiments, the step 310 includes pre-processing and processing the data. In some embodiments, the step 310 includes mining data using unsupervised derived modeling or data mining and context based/supervised learning modeling. In some embodiments, the step 310 includes using an automatically building text knowledge mining system that can receive data from various sources and use various methods to extract phrase information as relevant to a topic or subject, such as drilling applications. In some examples, pre-processing and processing the data includes mapping phrases in a multidimensional vector space with each vector comprising various aspects (e.g., dataset) of a drilling rig dataset including numerical and/or empirical datasets. Drilling rig dataset phrases will be generated and be relevant to, for example, a digital well plan, a digital rig plan, alarms, equipment, recipes relevant to various basin/formations, mud properties, the drilling metadata, etc. In some embodiments, phrases are classified using an automatic classification method or using existing classification information. Additionally, the weights between, or of, the phrases are trained based on relevancy to the well construction operations. In some embodiments, the weights between the phrases are trained by using the empirical information supported by numerical datasets relevant to well construction operations. In some embodiments, pre-processing and processing the data includes data cleansing by removing noisy data through filtering, etc.; data normalization, which includes aligning data to ensure that data skewness does not impact the model; data transformation to build a hypothesis, which includes a statistical transformation in understanding relationship in data; missing value treatment, which accounts for lacking data points; feature engineering, which includes selecting and creating new features that increase model accuracy; and imbalance data treatment. Labeling data is often essential for training text classifiers, but is also difficult to accomplish accurately, especially for complex and abstract concepts. In one embodiment, one or more generative large language models is used to produce labels and rationales for large-scale text analysis. In some embodiments, the generative language model is GPT-4 of OpenAI based in San Francisco, California. In other embodiments, the generative language model comprises or uses any one or more models of EleutherAI based in Washington, DC. In some embodiments, the model is or uses a RAG model, which is based on a Retrieve, Analyze, Generate ("RAG") framework. In some embodiments, this RAG framework is a method used in natural language processing (NLP) and AI models that combines elements of both retrieval-based and generative models for generating text and was developed by the team at EleutherAI. Some example uses of the RAG framework comprise tasks such as question answering, text generation, and dialogue systems. In some embodiments, the generative language model is deployed locally at the rig site on the EDGE infrastructure but in other embodiments the generative language model uses a Transformer Neural network ("TNN"), which is a type of deep learning model designed primarily for processing sequences of data, such as text or time-series data or Recurrent Neural network ("RNN") or Generative Adversarial Network ("GAN") or Convolutional Neural Network ("CNN") or Auto encoder or similar such models or hybrid models (e.g., physics based models that includes reduced order models, data physics, and reduced physics). In some embodiments, preprocessing of the data requires data engineering techniques to ensure that the data is standardized; labeled; and cleaned. In some embodiments, the data is stored in PostgreSQL developed by PostgreSQL Global Development Group of San Barbara, California. In some embodiments, the model 105 incorporates Llama Go and PyTorch for language modeling and deep learning tasks. In some embodiments, a user interface is accessible via standard web browsers, a RESTful API handles user queries and returns responses, a PostgreSQL database stores the data and is accessible via SQL queries, and the RAG model interfaces with Llama Go and PyTorch for model operations.

In some embodiments, the step 315 includes training and validating the model 105. In some embodiments, the model 105 includes an artificial intelligence ("AI") model. Example models may be or include any one or more of: the RAG model, a machine learning (“ML”) algorithm such as for example Category Boosting or CatBoost, and a random forest algorithm. In some examples, training the model comprises one or more selection techniques, such as for example filter methods, embedded methods, cross validation, holdout method, feature engineering, genetic algorithm, hyperparameter tuning, and lasso regression. In some embodiments, feature selection techniques improve the accuracy of the model 105 by reducing overfitting and improving the interpretability of the model. In some embodiments, the model 105 evolves and/or is re-trained with new data sets. In some embodiments, built-in feature importances is provided; built-in feature importances with random variable is provided; permutation importance is provided; and permutation importances with random variable are provided in relation to the model 105 to compare results. In some embodiments, training the model 105 at the step 315 includes improving vector search performance. In some embodiments, the model 105 is a Llama 3.1 by Meta of Menlo Park, California or other open source model. In some embodiments, the model 105 is a RAG system using Llama 3.1 or other open source model.

An Example Drilling Rig

Returning back to FIG. 1 and the apparatus 100, in some embodiments the apparatus 100 is or includes a land-based drilling rig. However, one or more aspects of the present disclosure are applicable or readily adaptable to any type of drilling rig, such as jack-up rigs, semisubmersibles, drill ships, coil tubing rigs, well service rigs adapted for drilling and/or re-entry operations, and casing drilling rigs, among others within the scope of the present disclosure. Moreover, the rig is not limited to a drilling rig but can also include a structure that is associated with activities performed prior to drilling activities and activities performed after drilling activities, such as completion operations.

Generally, the apparatus 100 monitors, in real-time, operations relating to a wellbore and optimizes instructions based on the real-time data. As used herein, the term “real-time” is thus meant to encompass close to real-time. As illustrated, the apparatus 100 includes a mast 100a supporting lifting gear above a rig floor 100b. The lifting gear includes a crown block 100c and a traveling block 100d. The crown block 100c is coupled at or near the top of the mast 100a, and the traveling block 100d hangs from the crown block 100c by a drilling line 100e. One end of the drilling line 100e extends from the lifting gear to draw works 100f, which is configured to reel out and reel in the drilling line 100e to cause the traveling block 100d to be lowered and raised relative to the rig floor 100b. The draw works 100f may include a rate of penetration (“ROP”) sensor, which is configured for detecting an ROP value or range, and a controller to feed-out and/or feed-in of a drilling line 100e. The other end of the drilling line 100e is anchored to a fixed position, possibly near the draw works 100f or elsewhere on the rig. In some embodiments, a drive system 100g is attached to the bottom of the traveling block 100d and a string 100h, such as a drill string or working string, extends from the top drive and is suspended within a wellbore 100i. In some embodiments, the drive system 100g is a top drive, but other drive systems, such as a power swivel, a rotary table, a coiled tubing unit, a downhole motor, and/or a conventional rotary rig, among others may be used.

The string 100h includes interconnected sections of drill pipe. When the string 100h is a drill string, a BHA 100j is attached. The BHA 100j may include one or more measurement-while-drilling (“MWD”) or wireline conveyed instruments, flexible connections, optional motors, adjustment mechanisms for push-the-bit drilling or bent housing and bent subs for point-the-bit drilling, a controller, stabilizers, and/or drill collars, among other components. When the string 100h is a working string, a completion system may be attached, a fishing tool may be attached, or other downhole tool attached. One or more pumps 100k may deliver drilling fluid to the string 100h via the top drive 100g.

The apparatus 100 also includes the computer system 100l that is operably coupled to a plurality of sensors, such as for example the ROP sensor, sensors that form a portion of the string 100h, the BHA 100j and/or a completion system that is attached to the string 100h, a surface casing annular pressure sensor configured to detect the pressure in the annulus defined between, for example, the wellbore 100i (or casing therein) and the string 100h, downhole annular pressure sensor coupled to or otherwise associated with the BHA 100j, a shock/vibration sensor that is configured for detecting shock and/or vibration in the BHA 100j, a mud motor delta pressure (ΔP) sensor that is configured to detect a pressure differential value or range across the one or more optional motors of the BHA 100j, one or more torque sensors, such as a bit torque sensor, may also be included in the BHA 100j for sending data to the computer system 100l that is indicative of the torque applied to the bit, a toolface sensor configured to estimate or detect the current toolface orientation or toolface angle, a WOB sensor integral to the BHA 100j and configured to detect WOB at or near the BHA 100j, an inclination sensor configured to detect inclination at or near the BHA 100j, an azimuth sensor configured to detect azimuth at or near the BHA 100j, a torque sensor coupled to or otherwise associated with the drive 100g, and a speed sensor configured to detect a value or range of the rotational speed of the drive 100g. The plurality of sensors may also include sensors placed at the surface or on the rig that measure operation of the mud pumps 100k, etc. In some embodiments, the computer system 100l is also operably coupled to a plurality of cameras that are positioned around or on the rig.

The detection performed by the sensors described herein may be performed once, continuously, periodically, and/or at random intervals. The detection may be manually triggered by an operator or other person accessing a human-machine interface (“HMI”) or GUI, or automatically triggered by, for example, a triggering characteristic or parameter satisfying a predetermined condition (e.g., expiration of a time period, drilling progress reaching a predetermined depth, drill bit usage reaching a predetermined amount, etc.). Such sensors and/or other detection means may include one or more interfaces which may be local at the well/rig site or located at another, remote location with a network link to the system.

Generally, the computing system 100l is configured to control or assist in the control of one or more components of the apparatus 100. For example, and as illustrated in the diagrammatic illustration 500 of portions of the system 50, the computing system 100l may be configured to receive data from a plurality of sensors, and using a front-end 105a of the trained model 105 and/or a local guidance application 505, transmit operational control signals to a variety of controllers, such as for example a mud pump(s) controller 510 that is in control of the mud pumps 100k, a drawworks controller 515 that is in control of the drive 100g, a downhole tool controller 520 that is in control of downhole tools, and other surface controllers 525 that are in control of other surface systems, such as for example speakers, cameras, etc. The computing system 100l may be a stand-alone component installed near the mast 100a and/or other components of the apparatus 100. In an example embodiment, the computing system 100l includes one or more systems located in a control room proximate the mast 100a, such as the general-purpose shelter often referred to as the “doghouse” serving as a combination tool shed, office, communications center, and general meeting place. The computing system 100l may be configured to transmit the operational control signals via wired or wireless transmission means which, for the sake of clarity, are not depicted in FIG. 1.

Generally, the computing system 100l is coupled to a graphical user interface (“GUI”) 530 that may include an input mechanism 530a and a display 530b. The computing system 100l generally includes a processor and a computer readable medium operably coupled thereto. Instructions accessible to, and executable by, the computer processor are stored on the computer readable medium. A database is also stored in the computer readable medium. Generally, the GUI 530 can display a plurality of windows or screens to the user. In some embodiments, the user provides inputs to the system 50 via a window that is displayed on the GUI 530a. As illustrated in FIG. 5, the front-end 105a of a trained model 105 may be stored in the computing system 100l via an application but in some embodiments the trained model 105 is accessible to the computing system 100l via the network 110 and a front-end 105a of the trained model is not stored in the computing system 100l. In some embodiments, the local guidance application 505 is stored in or otherwise accessible to the computing system 100l. In some embodiments, the local guidance application 505 (e.g., “EDGE”) is configured to process and respond to measured parameters and conditions in a real-time or near-real-time basis.

In some embodiments, each of the local guidance application 505 and the front-end 105a is or includes an application stored in the computer readable medium of the computing system 100l. Another portion of each of the local guidance application 505 and the front-end 105a may be stored in the cloud or on a server that is remote from the computing system 100l.

In some embodiments, the local guidance application 505 and the trained model 105 are in communication and working in combination. For example, signals, data, and/or instructions can be sent from the local guidance application 505 to the front-end 105a of the trained model 105 and vice versa.

An Example Method of Using the System

In some embodiments, and as illustrated in the flow diagram of FIG. 6, a method 600 of using the trained model 105 comprises accessing input data at step 605; pre-processing and processing the data at step 610; generating, by the trained model, output at step 615; and controlling a portion of the rig apparatus 100 using the generated output at step 620.

In some embodiments, the step 605 includes accessing input data 230, which is a portion of the data 205. Generally, at least a portion of the input data 230 is accessed in real-time or near real-time. In some embodiments, the input data 230 includes data from the plurality of sensors associated with the apparatus 100. In some examples, the input data 230 includes surface data 205g, such as data from a camera positioned to capture activity in a portion of the apparatus 100; operations data 205j that includes data regarding the operations currently being performed (e.g., drilling, taking a survey, tripping out); and rig inventory data 205d for the apparatus 100. In some embodiments, the data from the camera may include visuals or frames that form a video.

In some embodiments, the step 610 includes pre-processing and processing the data. In some embodiments, the step 610 is substantially similar to the step 310 except that the step 610 is pre-processing and processing the input data 230 instead of the training data 220 and validation data 225.

In some embodiments, the step 615 includes generating, by the trained model 105, an output 215. In some embodiments, generating output by the trained model 105, includes identifying activities in real-time and/or in near real-time; predicting, based on historical data and the identified activities, a future rig activity; and generating an output 215, such as a recommendation, advisory, and/or alarm based on the predicted future activity. For example, and when the input data 230 includes a camera capturing a worker in a space associated with the apparatus 100, the trained model 105 may identify the worker based on: facial recognition technology; based on an identification badge worn on the clothing or helmet; and/or based on a signal or other output from an electronic device that is carried or worn by the worker. In some embodiments, worker data associated with the worker, such as the name, the position, the training or other licensing completed, is identified by the trained model 105 using, for example, the human resources data 205i. The trained model 105 also identifies the location and/or trajectory of the worker, for example, the worker may be walking through or toward an area of the apparatus 100 that is a restricted area or an area that is reserved for a specifically trained worker. The trained model 105 also identifies the operations being performed by the apparatus 100, using for example the operations data 205j, and the equipment used in those operations, using for example the rig inventory data 205d and the operations data 205j. The trained model 105, after identifying the worker, the worker data, the location and/or trajectory of the worker, and the operations of the apparatus 100, makes a prediction of the future activity of the worker. For example, when the worker is walking toward an area of the apparatus 100 that is a restricted area, the prediction may be that the worker is going to enter the restricted area. In some embodiments, the prediction is based on historical data of other worker activities. In some embodiments, the area being a restricted area is based on the operations performed by the apparatus 100. For example, the area surrounding and including a mousehole formed in the rig floor may not be a restricted area when the operations involve a pause on placing and/or removing pipe sections from the mousehole. However, the same location (i.e., the area surrounding and including the mousehole) may be a restricted area for workers to walk around when drilling is commenced, and a section of pipe is expected to be placed into the mousehole at a regular interval. In the example where the trained model 105 predicts the worker to be walking toward the restricted area, the trained model 105 may generate an output 215 that includes a prediction that the worker is at an increased risk of being struck by a pipe and/or an output 215 recommending equipment that is controlling the movement of pipe to/from the mousehole be paused. Other output 215 may include generating a visual, audible, or haptic alarm in the area local to the worker and/or other locations.

In some embodiments, the step 620 includes controlling a portion of the rig apparatus 100 using the generated output 215. For example, the model 105, based on the output 215, controls or instructs the controllers 510, 515, 520 and/or 525 to pause the movement of the pipe to/from the mousehole and/or generate the recommended alarm. In some embodiments, an audible output is generated at the step 620 via speakers that form a portion of the rig. In some embodiments, the model 105 communicates to the computing system 100l, including camera systems and the digital twin, and directly or indirectly provides advisory and actions to the rig controller for process automation and other modules to manage the operations of a drilling rig.

Example Alternatives and Advantages of the System

While the output 215 detailed above relates to a local (e.g., on or near the apparatus 100) response that occurs in real-time or near real-time, in some embodiments, the output 215 is based on the monitored activities at the apparatus 100 over a longer period of time. For example, the inputs or input data 230 may be data detailing activities monitored by the model 105 during one month and the output 215 includes a report of the most frequent or most critical activities occurring in that month. In response, the output 215 of the model 105 may further include customized training or provide some other advisory specific to the apparatus 100 to reduce the frequency or criticality of the activities in the future. In some embodiments, and when the model 105 reviews the data to identify the highest cause of incident over a period of time, the model 105 may identify that the years of experience, strain/overexertion, and circadian rhythm (e.g., incidents between 6 am to 12 pm) are correlated with the highest causes of accidents. In some embodiments, the cause of an incident is or relates to a risk factor. For example, when circadian rhythm is identified as correlating with a high number of incidents, then the operations occurring between 6 am to 12 pm, may be considered higher risk. For example, when overexertion is identified as correlating with a high number of incidents, then shifts that are greater than 9 hours or the last three hours of each 12 hour shift may be considered higher risk. In some embodiments, the output 215 of the model 105 includes an incident summary per a period of time (e.g., one month) and an advisory based on the incident summary (e.g., creation of instruction videos configured to reduce or eliminate a portion of the incidents). For example, the summary could identify that the role of motorhand and floorhand have the highest number of incidents while drilling during rigging up and down; that most incidents have occurred in North Dakota; and that the mechanisms that need to be reviewed are strain/overexertion and “stuck by.” In response, the model 105 may generate an advisory that includes: content driving safety conversations and safety audits in that area; driver safety videos related to mechanism; and a recommendation for circadian rhythm management. In some embodiments, the model 105 may generate an output 215 that comprises a recommendation specific to the risk factor and configured to mitigate the incident type (e.g., rigging up incidents, rigging down incidents). In some embodiments, the recommendation includes training materials customized to the risk factor; and wherein the trained model 105 generates the customized training materials.

While the output 215 detailed above relates to the prevention of an injury, the output 215 may also relate to instructional advisory or outputs. For example, instead of identifying the worker walking into a restricted area, the system 50 may identify that the worker is predicted to pick up an item and the output 215 may include an audible instruction on how to best pick up the item. As such, the model 105 may understand current behavior of the worker and the output 215 may include optimum ergonomics or behavior. That is, in some embodiments, the output 215 is configured to optimize the performance and/or behavior.

Another example output 215 relates to recipes, formation, and/or well-related information. For example, the output 215 may include an alert relating to the location of equipment that is predicted to be needed for an upcoming section of the well. Other types of electronic alerts and advisories may be included. In some embodiments, the system 50 and/or the model 105 uses historical modeling with physics informed AI to provide context to formation tops and formation properties; builds contextual recipe datasets based on the formation tops/historical datasets; auto-generates well-related advisories based on formation tops/historical datasets. As such, and in some embodiments, the input data includes formation data generated during drilling activities, the risk factor identified by the trained model 105 includes a formation-related risk factor (e.g., less than ideal drilling conditions or incidents due to reaching a formation top earlier or later than expected), and the output includes a formation related prediction (e.g., when a formation top will be reached, what type of formation top will be reached) and target drilling instructions based on the formation related prediction.

In some embodiments, a physics based deep learning system is integrated in the model 105 and/or the system 50 such that the model 105 is a generative model that automatically produces all the necessary files to run a simulation in an open-source code, complete with computational mesh, and integrate data associated with the simulation into the solution, which can be used to enhance the digital twin running on the drilling rig 100 for enhanced decision making and also for enhanced process automation.

In some embodiments, the system 50 and/or the model 105 receives a query from a requestor and forms a closed loop system such that contextual information associated with the requestor is determined in relation to the query. In some embodiments, the system 50 and/or the model 105 queries a knowledge graph with the query and the contextual information. An answer, including a source in some embodiments, associated with the query is identified within the knowledge graph by the system 50 and/or the model 105. The answer is then provided to the requestor by the system 50 and/or the model 105. Upon the system 50 and/or the model 105 receiving feedback associated with the answer from the requestor, the knowledge graph is modified based on the feedback.

In some embodiments, the system 50 and/or the model 105 provides an improvement to the technical field of well construction operations because the system 50 and/or the model 105 enables knowledge sharing via the collection, conversion, and consolidation of data from various formats, locations, system, application, etc. into a knowledge mining system having a standardized format and using that knowledge mining system to generate and implements outputs based on real-time, near real-time, or recent trends of activities on a drilling rig.

In some embodiments, the system 50 and/or the model 105 labels alarms; labels optimum responses to alarms; responds or takes action to messages and non-critical alarms on occurrence (if required); and/or provides an advisory to a driller on critical alarms.  In some embodiments, the system 50 and/or the model 105 takes actions such as inserting tickets, notifying required personnel, dispatching technicians etc.

In some embodiments, the system 50 and/or the model 105 labels cause and resolution from historical maintenance problems and aiding technicians. In some embodiments, aiding or assisting technicians can include advisory in terms of schematics, resolution, paperwork assistance (via voice commands) etc.

In some embodiments, the system 50 and/or the model 105 improves decision making, optimizes operations, enhances customer experience, and creates new products and services. In some embodiments, the system 50 and/or the model 105 provides an improvement to the technical field of well construction operations because the system 50 and/or the model 105: reduces costs and increases efficiency by automating tasks that require human intelligence, such as data analysis, classification, summarization, and visualization; improves accuracy and reliability by eliminating human errors, biases, and inconsistencies in data processing and interpretation; uncovers hidden patterns and trends that are not obvious or accessible to human experts, such as customer preferences, market opportunities, and risk factors; enables innovation and creativity by generating new ideas, hypotheses, and solutions based on data-driven insights; and/or enhances collaboration and communication by facilitating the sharing and dissemination of knowledge across different teams, departments, and stakeholders.

Other Considerations

In an example embodiment, the network 110 includes the Internet, one or more local area networks, one or more wide area networks, one or more cellular networks, one or more wireless networks, one or more voice networks, one or more data networks, one or more communication systems, and/or any combination thereof.

As used herein, the term “substantially the same” can be understood to mean similar historical conditions likely to lead to the same result in the present, e.g., based on a similar geologic formation and the same drilling conditions or the same geologic formation and similar drilling conditions, or the like. In the event the above wording is insufficiently precise, the term “similar,” “similar to,” or “substantially the same” could also be understood herein to mean current numerical values that are up to about ten percent (10%) above or below the historical data, or historical data which are up to about ten percent (10%) above or below the current condition.

Methods within the scope of the present disclosure may be local or remote in nature. These methods, and any controllers discussed herein, may be achieved by one or more intelligent adaptive controllers, programmable logic controllers, artificial neural networks, and/or other adaptive and/or “learning” controllers or processing apparatus. For example, such methods may be deployed or performed via PLC, PAC, PC, one or more servers, desktops, handhelds, and/or any other form or type of computing device with appropriate capability.

In an example embodiment, as illustrated in FIG. 7 with continuing reference to FIGS. 1-6, an illustrative node 1000 for implementing one or more of the example embodiments described above and/or illustrated in FIGS. 1-6 is depicted. The node 1000 includes a microprocessor 1000a, an input device 1000b, a storage device 1000c, a video controller 1000d, a system memory 1000e, a display 1000f, and a communication device 1000g, all interconnected by one or more buses 1000h. In several example embodiments, the storage device 1000c may include a floppy drive, hard drive, CD-ROM, optical drive, any other form of storage device and/or any combination thereof. In several example embodiments, the storage device 1000c may include, and/or be capable of receiving, a floppy disk, CD-ROM, DVD-ROM, or any other form of computer-readable medium that may contain executable instructions. In several example embodiments, the communication device 1000g may include a modem, network card, or any other device to enable the node to communicate with other nodes. In several example embodiments, any node represents a plurality of interconnected (whether by intranet or Internet) computer systems, including without limitation, personal computers, mainframes, PDAs, smartphones, and cell phones.

In several example embodiments, one or more of the components of the systems described above and/or illustrated in FIGS. 1-6 include at least the node 1000 and/or components thereof, and/or one or more nodes that are substantially similar to the node 1000 and/or components thereof. In several example embodiments, one or more of the above-described components of the node 1000, the system 10, and/or the example embodiments described above and/or illustrated in FIGS. 1-6 include respective pluralities of same components.

In several example embodiments, one or more of the applications, systems, and application programs described above and/or illustrated in FIGS. 1-6 include a computer program that includes a plurality of instructions, data, and/or any combination thereof; an application written in, for example, Arena, Hypertext Markup Language (HTML), Cascading Style Sheets (CSS), JavaScript, Extensible Markup Language (XML), asynchronous JavaScript and XML (Ajax), and/or any combination thereof; a web-based application written in, for example, Java or Adobe Flex, which in several example embodiments pulls real-time information from one or more servers, automatically refreshing with latest information at a predetermined time increment; or any combination thereof.

In several example embodiments, a computer system typically includes at least hardware capable of executing machine readable instructions, as well as the software for executing acts (typically machine-readable instructions) that produce a desired result. In several example embodiments, a computer system may include hybrids of hardware and software, as well as computer sub-systems.

In several example embodiments, hardware generally includes at least processor-capable platforms, such as client-machines (also known as personal computers or servers), and hand-held processing devices (such as smart phones, tablet computers, personal digital assistants (PDAs), or personal computing devices (PCDs), for example). In several example embodiments, hardware may include any physical device that is capable of storing machine-readable instructions, such as memory or other data storage devices. In several example embodiments, other forms of hardware include hardware sub-systems, including transfer devices such as modems, modem cards, ports, and port cards, for example.

In several example embodiments, software includes any machine code stored in any memory medium, such as RAM or ROM, and machine code stored on other devices (such as floppy disks, flash memory, or a CD ROM, for example). In several example embodiments, software may include source or object code. In several example embodiments, software encompasses any set of instructions capable of being executed on a node such as, for example, on a client machine or server.

In several example embodiments, combinations of software and hardware could also be used for providing enhanced functionality and performance for certain embodiments of the present disclosure. In an example embodiment, software functions may be directly manufactured into a silicon chip. Accordingly, it should be understood that combinations of hardware and software are also included within the definition of a computer system and are thus envisioned by the present disclosure as possible equivalent structures and equivalent methods.

In several example embodiments, computer readable mediums include, for example, passive data storage, such as a random-access memory (RAM) as well as semi-permanent data storage such as a compact disk read only memory (CD-ROM). One or more example embodiments of the present disclosure may be embodied in the RAM of a computer to transform a standard computer into a new specific computing machine. In several example embodiments, data structures are defined organizations of data that may enable an embodiment of the present disclosure. In an example embodiment, a data structure may provide an organization of data, or an organization of executable code.

In several example embodiments, any networks and/or one or more portions thereof may be designed to work on any specific architecture. In an example embodiment, one or more portions of any networks may be executed on a single computer, local area networks, client-server networks, wide area networks, internets, hand-held and other portable and wireless devices, and networks.

The present disclosure introduces a method of building a well construction knowledge mining system using an artificial intelligence ("AI") model and a plurality of databases, the method comprising: accessing, within the plurality of databases, a set of well construction files; wherein the set of well construction files comprise files in a plurality of formats; and wherein the set of well construction files comprise data relating to well construction operations; preprocessing the set of well construction files to generate a training data set, comprising: reformatting at least a portion of the data to standardized machine-readable data; labeling data in the machine-readable files using the model; and cleaning the data; and training the model, using supervised and unsupervised learning and the training data set, thereby building the well construction knowledge mining system. In some embodiments, the method also includes receiving well construction data associated with a rig activity; providing, as input to the trained model, the well construction data; predicting, by the trained model and based on the well construction data, a future rig activity; and generating, by the trained model, output that is directed to the future rig activity. In some embodiments, the method also includes comprising automatically controlling, by a controller of a rig and in response to the output, at least a portion of the rig based on the output. In some embodiments, the output comprises an advisory relating to the future rig activity. In some embodiments, the advisory comprises a recommended action. In some embodiments, the data relating to well construction operations comprises: data generated in response to a safety incident involving a personnel associated with a drilling rig; and data that identifies: the role of the personnel, the area where the incident occurred, and whether an external parameter is associated with the safety incident. In some embodiments, the data relating to well construction operations comprises data generated in response a maintenance task associated with equipment; wherein the future rig activity is associated with rig equipment; and wherein the output that is directed to the future rig activity comprises a recommendation to perform maintenance on the rig equipment. In some embodiments, the set of well construction files comprises: data generated in response a maintenance task associated with equipment; and data generated to record equipment failure and data associated with a recording system and run-time logs from a drilling rig. In some embodiments, the rig activity comprises a drilling activity; wherein the well construction data comprises downhole data; wherein the downhole data comprises formation data generated during the drilling activities; and wherein the output comprises: a formation related prediction; and target drilling instructions based on the formation related prediction. In some embodiments, the model comprises a retrieval-augmented generation model.

The present disclosure also introduces a system configured to build a well construction knowledge mining system using an artificial intelligence ("AI") model and a plurality of databases, the system comprising a non-transitory computer readable medium having stored thereon a plurality of instructions, wherein the instructions are executed with one or more processors so that the following steps are executed: accessing, within the plurality of databases, a set of well construction files; wherein the set of well construction files comprise files in a plurality of formats; and wherein the set of well construction files comprise data relating to well construction operations; preprocessing the set of well construction files to generate a training data set, comprising: reformatting at least a portion of the data to standardized machine-readable data; labeling data in the machine-readable files using the model; and cleaning the data; and training the model, using supervised and unsupervised learning and the training data set, thereby building the well construction knowledge mining system. In some embodiments, wherein the instructions are executed with one or more processors so that the following steps are executed: receiving well construction data associated with a rig activity; providing, as input to the trained model, the well construction data; predicting, by the trained model and based on the well construction data, a future rig activity; and generating, by the trained model, output that is directed to the future rig activity. In some embodiments, the instructions are executed with one or more processors so that the following step is executed: automatically controlling, by a controller of a rig and in response to the output, at least a portion of the rig based on the output. In some embodiments, the output comprises an advisory relating to the future rig activity. In some embodiments, the advisory comprises a recommended action. In some embodiments, the data relating to well construction operations comprises: data generated in response to a safety incident involving a personnel associated with a drilling rig; and data that identifies: the role of the personnel, the area where the incident occurred, and whether an external parameter is associated with the safety incident. In some embodiments, the data relating to well construction operations comprises data generated in response a maintenance task associated with equipment; wherein the future rig activity is associated with rig equipment; and wherein the output that is directed to the future rig activity comprises a recommendation to perform maintenance on the rig equipment. In some embodiments, the set of well construction files comprises: data generated in response a maintenance task associated with equipment; and data generated to record equipment failure and data associated with a recording system and run-time logs from a drilling rig. In some embodiments, the rig activity comprises a drilling activity; wherein the well construction data comprises downhole data; wherein the downhole data comprises formation data generated during the drilling activities; and wherein the output comprises: a formation related prediction; and target drilling instructions based on the formation related prediction. In some embodiments, the model comprises a retrieval-augmented generation model.

In several example embodiments, a database may be any standard or proprietary database software. In several example embodiments, the database may have fields, records, data, and other database elements that may be associated through database specific software. In several example embodiments, data may be mapped. In several example embodiments, mapping is the process of associating one data entry with another data entry. In an example embodiment, the data contained in the location of a character file can be mapped to a field in a second table. In several example embodiments, the physical location of the database is not limiting, and the database may be distributed. In an example embodiment, the database may exist remotely from the server, and run on a separate platform. In an example embodiment, the database may be accessible across the Internet. In several example embodiments, more than one database may be implemented.

In several example embodiments, a plurality of instructions is stored on a non-transitory computer readable medium, and the instructions are executed by one or more processors to cause the one or more processors to carry out or implement in whole or in part the above-described operation of each of the above-described example embodiments of the system, the method, or any combination thereof. In several example embodiments, such a processor may include one or more of the microprocessor 1000a, any processor(s) that are part of the components of the system, and/or any combination thereof, and such a non-transitory computer readable medium may be distributed among one or more components of the system. In several example embodiments, such a processor may execute the plurality of instructions in connection with a virtual computer system. In several example embodiments, such a plurality of instructions may communicate directly with the one or more processors, and/or may interact with one or more operating systems, middleware, firmware, other applications, and/or any combination thereof, to cause the one or more processors to execute the instructions.

The term “about,” as used herein, should generally be understood to refer to both numbers in a range of numerals. For example, “about 1 to 2” should be understood as “about 1 to about 2.” Moreover, all numerical ranges herein should be understood to include each whole integer, or 1/10 of an integer, within the range.

The phrase “at least one of A and B” should be understood to mean “A; B; or both A and B.”  The phrase “one or more of the following: A, B, and C” should be understood to mean “A; B; C; A and B; B and C; A and C; or all three of A, B, and C.” The phrase “one or more of A, B, and C” should be understood to mean “A; B; C; A and B; B and C; A and C; or all three of A, B, and C.”

In several example embodiments, while different steps, processes, and procedures are described as appearing as distinct acts, one or more of the steps, one or more of the processes, and/or one or more of the procedures could also be performed in different orders, simultaneously and/or sequentially. In several example embodiments, the steps, processes and/or procedures could be merged into one or more steps, processes and/or procedures.

It is understood that variations may be made in the foregoing without departing from the scope of the disclosure. Furthermore, the elements and teachings of the various illustrative example embodiments may be combined in whole or in part in some or all of the illustrative example embodiments. In addition, one or more of the elements and teachings of the various illustrative example embodiments may be omitted, at least in part, and/or combined, at least in part, with one or more of the other elements and teachings of the various illustrative embodiments.

Any spatial references such as, for example, “upper,” “lower,” “above,” “below,” “between,” “vertical,” “horizontal,” “angular,” “upwards,” “downwards,” “side-to-side,” “left-to-right,” “right-to-left,” “top-to-bottom,” “bottom-to-top,” “top,” “bottom,” “bottom-up,” “top-down,” “front-to-back,” etc., are for the purpose of illustration only and do not limit the specific orientation or location of the structure described above.

In several example embodiments, one or more of the operational steps in each embodiment may be omitted or rearranged. Moreover, in some instances, some features of the present disclosure may be employed without a corresponding use of the other features. Moreover, one or more of the above-described embodiments and/or variations may be combined in whole or in part with any one or more of the other above-described embodiments and/or variations.

Although several example embodiments have been described in detail above, the embodiments described are examples only and are not limiting, and those skilled in the art will readily appreciate that many other modifications, changes, and/or substitutions are possible in the example embodiments without materially departing from the novel teachings and advantages of the present disclosure. Accordingly, all such modifications, changes, and/or substitutions are intended to be included within the scope of this disclosure as defined in the following claims. In the claims, any means-plus-function clauses are intended to cover the structures described herein as performing the recited function and not only structural equivalents, but also equivalent structures. Moreover, it is the express intention of the applicant not to invoke 35 U.S.C. § 112(f) for any limitations of any of the claims herein, except for those in which the claim expressly uses the word “means” together with an associated function.