METHOD AND MACHINE-READABLE MEDIUM FOR DATA-CENTRIC DRILLING HAZARD PREDICTION

Abstract

A method includes inputting a dataset including data about a wellbore, a geographical area, a hydrocarbon formation, or any combination thereof, to one or more artificial intelligence models, generating a prediction, via the one or more artificial intelligence models, of a probability of a hazard event for the wellbore, an impending hazard event during active drilling of the wellbore, an optimal location for the wellbore in the geographical area, or any combination thereof, and performing or modifying drilling operations in response to the prediction generated by the one or more artificial intelligence models.

Description

FIELD OF THE DISCLOSURE

The present disclosure relates generally to planning and constructing oil and gas wells and, more particularly, to utilizing a data-centric artificial intelligence approach for the prediction of drilling hazards.

BACKGROUND OF THE DISCLOSURE

During the drilling of oil and gas wells, several factors may create drilling hazards such as stuck drilling pipe or a loss of drilling fluid circulation, which often lead to the accumulation of non-productive time (NPT) and costly corrective measures. In order to reduce the occurrence of NPT and corrective measures in the construction of an oil and gas well, the specific factors which lead to these operational issues must be properly understood. These factors may include the natural characteristics of the subterranean formations, parameters of the drilling process, or the fracturing that has occurred within the formation. The ultimate cause of the drilling hazards within the subterranean formation, however, is often a combination of multiple, independent factors which may combine to create unfavorable drilling conditions.

Accordingly, a method for the tracking, combination, and correlation of these factors is desirable for enhanced operation and planning of oil and gas wells.

SUMMARY OF THE DISCLOSURE

Various details of the present disclosure are hereinafter summarized to provide a basic understanding. This summary is not an exhaustive overview of the disclosure and is neither intended to identify certain elements of the disclosure, nor to delineate the scope thereof. Rather, the primary purpose of this summary is to present some concepts of the disclosure in a simplified form prior to the more detailed description that is presented hereinafter.

According to an embodiment consistent with the present disclosure, a method includes inputting a dataset comprising data about a wellbore, a geographical area, a hydrocarbon formation, or any combination thereof, to one or more artificial intelligence models, generating a prediction, via the one or more artificial intelligence models, of a probability of a hazard event for the wellbore, an impending hazard event during active drilling of the wellbore, an optimal location for the wellbore in the geographical area, or any combination thereof, and performing or modifying drilling operations in response to the prediction generated by the one or more artificial intelligence models.

In a further embodiment, a system includes a data processing application comprising a data mining module and a data filter module for receiving, extracting, and filtering a dataset, an artificial intelligence model training application comprising one or more machine learning training modules, and a hazard prediction application configured to predict a hazard event within a wellbore utilizing one or more artificial intelligence models.

Any combinations of the various embodiments and implementations disclosed herein can be used in a further embodiment, consistent with the disclosure. These and other aspects and features can be appreciated from the following description of certain embodiments presented herein in accordance with the disclosure and the accompanying drawings and claims.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 is a flowchart of an example method for the data preprocessing, training and application of an artificial intelligence model for drilling hazard predictions.

FIG. 2 is a schematic of an example system for the data preprocessing, training and application of an AI model for drilling hazard prediction.

FIG. 3 illustrates one example of a computer system that can be employed to execute one or more embodiments of the present disclosure.

DETAILED DESCRIPTION

Embodiments of the present disclosure will now be described in detail with reference to the accompanying Figures. Like elements in the various figures may be denoted by like reference numerals for consistency. Further, in the following detailed description of embodiments of the present disclosure, numerous specific details are set forth in order to provide a more thorough understanding of the claimed subject matter. However, it will be apparent to one of ordinary skill in the art that the embodiments disclosed herein may be practiced without these specific details. In other instances, well-known features have not been described in detail to avoid unnecessarily complicating the description. Additionally, it will be apparent to one of ordinary skill in the art that the scale of the elements presented in the accompanying Figures may vary without departing from the scope of the present disclosure.

Embodiments in accordance with the present disclosure generally relate to planning and constructing oil and gas wells and, more particularly, to utilizing a data-centric artificial intelligence approach for the prediction of drilling hazards. Through the accumulation of data from drilling crew reports, static well characteristics, and real-time sensors, artificial intelligence models may be trained and tested to predict well hazards prior to their occurrence. The artificial intelligence models may be further used for the determination of optimal well design or construction location and trajectory in order to produce efficient wells in low-risk locations. These enhanced operations provided by the artificial intelligence models described herein may reduce drilling costs, non-productive time, and environmental damage while improving the health, safety, and environment (HSE) ratings of the oil and gas wells.

FIG. 1 is a flowchart of an example method 100 for the training and application of an artificial intelligence (AI) model for drilling hazard predictions. The method 100 may begin at 102 with the collection of data from each available source. The data which may be collected at 102 may include, but is not limited to, one or more drilling reports detailing the drilling hazard from the drilling operation crew, the static wellbore data, and the real-time sensor data from the time period before, during, and after the event with timestamped entries. The one or more drilling reports may include a description of the event, the unique well and activity identifiers defined within the reporting system, and additional metadata related to the drilling hazard. In some embodiments, the static wellbore data may include the well trajectory including the azimuth and inclination, the formation top depth, the properties of the drilling fluid utilized, the depth of the wellbore, and any other known attributes of the well or the drilling operation. Similarly, the real-time sensor data may include, but is not limited to depth measurement at the time of the drilling hazard, the weight of the drilling string (hook-load), rate of penetration (drilling speed), weight on the drill bit, torque measurements, pressure measurements, drilling fluid tank level, and any other actively monitored drilling condition.

The data collected at 102 may be mined and transformed at 104, 106, and 108 depending on the data type collected at 102. The one or more drilling reports detailing the drilling hazard from the drilling operation crew may be transformed at 104 through the use of text mining, in which regular expressions are utilized to perform the search in all historical data for all the wells available in the database. Natural language processing (NLP) algorithms may be further employed at 104 in order to extract contextual information and refine the results to include the key points of the reports. Additionally, NLP may be used to detect obscure hazardous events based on sentiment not direct wording reported by the drilling crew. The transformation at 104 may output a list of hazard events based upon human comments which includes event details and metadata from the wellbore of interest.

The static wellbore data may be parsed at 106 to extract the relevant information for the wellbore of interest. In some embodiments, the list of hazard events output from 104 may be provided as an input at 106. Through the input of the list of hazard events, the static wellbore data may be paired to each wellbore for each event on the list of hazard events. The paired wellbore data may include the wellbore trajectory, the formation name and top depths, the drilling fluid properties at the time of the hazard event, and additional information that may be dependent on the hazard type defined by the input list. Any parameters previously extracted which were not sampled along the depth of the wellbore may be further transformed at 106 in order to be indexed by depth for further analysis.

The drilling sensor data may be transformed at 108 in order to fully construct the initial dataset. The drilling sensor data may first be transformed to be indexed by depth (time domain to spatial domain transformation) to correlate to the further data types transformed at 104 and 106. Similar to the transformation at 106, the transformation at 108 may receive the outputs of 104 and 106 as inputs in order to better correlate the readings and data to specific events in specific wellbores. The sensor data and the additional inputs may be used to determine the status of the drilling operation such that depth values may be altered if the drill was advancing or retreating during the event at issue. The drilling sensor data may further include well logs and seismic data which may be similarly correlated and paired to the data transformed at 104 and 106.

The outputs of 104, 106, and 108 may be combined at 110 to generate a preliminary dataset through preprocessing and filtering. The dataset created and preprocessed at 110 may be a combination of static formation characteristics and sensor readings correlated to specific events within specific wellbores. The sensor readings may be specifically preprocessed and filtered to search for a trend which may indicate the hazard event reported by the drilling crew. By way of non-limiting example, in a loss of circulation event the measurements in the drilling fluid tank level and the stand pipe pressure will change as a result of the circulation issues. Moreover, changes in drilling speed, weight-on-bit (WOB) might also be seen due to root-causes of circulation issues. However, without the filters applied at 110, the changes in measurements may be undetectable within the datasets. As such, the filters may be designed to detect changes within a certain frequency range and filter out expected changes due to planned drilling operations.

With the data pre-processed at 110, the depth at which the event was detected may be flagged after being inferred from the sensor data, and the severity of the event may be quantified based upon the rate of change of the sensor measurements at 112. The flagging and severity quantification of the event at 112 may be included within the preliminary dataset from 110 and may be compiled to confirm the hazard event and correct any errors in the inferred depth range or event severity to generate a high-quality dataset at 114 for each wellbore which have been indexed by depth and may include a configurable resolution (e.g., a depth resolution of 1 foot).

The high-quality dataset generated at 114 may be provided to one or more AI models 115 at 116 for training purposes, such that the AI models receive the high-quality dataset as the input to generate the output hazard event which has been flagged and quantified. The one or more AI or machine learning training techniques may include, but are not limited to, Logistic Regression (LR), Naïve Beyes (NB), K-Nearest Neighbor (KNN), Decision Tree (DT), Ada Boost (AB), Deep Neural Network (DNN), Support Vector Machine (SVM), or Random Forest (RF). Further, ensemble machine learning techniques may also be used. These machine learning techniques may be used for the training of the one or more AI models at 116 such that the correlation between the components of the high-quality dataset and the hazard event outcome may be determined.

Following the training of the one or more AI models at 116, the one or more AI models 115 may be employed at 118, 120, and 122. At 118, the one or more AI models may receive either raw or pre-processed data regarding a planned or constructed well including the static wellbore data previously discussed as well as the geological information about the subterranean formation or the surrounding area. The one or more AI models may utilize the trained correlations at 118 to predict the occurrence or possibility of a hazard event during initial or further construction of the wellbore. The likelihood of a hazard event may be provided as an output by the one or more AI models at 118, as well as the location or compounding factors which may lead to the hazard event.

Similarly, the one or more AI models 115 may be provided the raw or pre-processed data regarding the wellbore while simultaneously receiving real-time measurements from one or more surface and downhole sensors in an active drilling operation at 120. The one or more AI models may monitor the input real-time measurements to determine a possible or impending hazard event based upon the correlations previously determined. The one or more AI models may provide a warning to a drill crew of the event, or may automatically adjust or cease drilling operations through communication with the drilling equipment to avert the hazard event.

The one or more AI models may additionally be provided the geography and seismic readings of a subterranean formation at 122. The subterranean formation may have been previously identified and flagged for drilling; however, the exact location of the wellbore may not have been previously defined. As such, the one or more AI models 115 may receive the geographical and seismic readings as an input to provide the optimal locations and trajectory for wellbore drilling in the region to avoid hazard events. Based upon the correlations previously defined, the problematic geographical and geological characteristics or features may be avoided and more favorable drilling locations and trajectory may be determined at 122.

For the one or more AI models to be improved upon after operation, the method 100 may further include a feedback loop at 124. The feedback loop may receive a high-quality dataset from the outcome of any drilling undertaken at 118, 120, 122 or any combination thereof. From the high-quality dataset and the successful avoidance or failed determination of a hazard event, the results may be provided back to the training techniques at 116 as positive or negative training sets for the one or more AI models 115.

FIG. 2 is a schematic of an example system 200 for the training and application of an AI model for drilling hazard prediction. The system 200 may include a network interface 202 which may enable the transfer or receipt of one or more files which may be utilized in the training or operation of one or more AI models. The files sent or received may include, but are not limited to one or more drilling reports detailing the drilling hazard from the drilling operation crew, static wellbore data detailing the geographical and geological characteristics of the wellbore, and real-time sensor data from the time period before, during, and after a drilling hazard event with timestamped entries. The network interface 202 may enable communication between the system 200 and any active drilling equipment which may be performing drilling operations, such that the system 200 may remotely alter or control the operation of the equipment via the network interface 202.

Further, the network interface 202 may communicate with one or more sensors 204, in instances where the one or more sensors 204 exist outside of the system and are network enabled. In some embodiments, however, the one or more sensors 204 may be directly connected within the system 204 such that a local bus connects the one or more sensors 204 with the remainder of the system 200. The one or more sensors may include, but are not limited to pressure transducers, torque transducers, weight transducers, fluid level transducers, and depth sensors, temperature sensors, and so on.

The system 200 may further include a database 206 which may receive and store data that may include, but is not limited to one or more drilling reports detailing the drilling hazard from the drilling operation crew, static wellbore data detailing the geographical and geological characteristics of the wellbore, and real-time sensor data from the time period before, during, and after a drilling hazard event with timestamped entries. The data stored within the database 206 may be pre-loaded, may be received via the network interface 202, or may be written to the database via the one or more sensors 204.

Connected to the various components outlined above, a processor 208 may enable the operation of the training and application of the one or more AI models. In some embodiments, the processor 208 may contain three separate applications which may communicate internally including a data processing application 210, an AI model training application 212, and a hazard prediction application 214.

The data processing application 210 may receive one or more data types from the network interface 202, the one or more sensors 204, the database 206, or any combination thereof. With the one or more datatypes, a data mining module 216 may receive the one or more datatypes and extract pertinent information through text mining and NLP algorithms. The one or more datatypes may be further indexed by depth and correlated to specific depths and events. The mined data may be passed to the data filter module 218 which may preprocess and filter the data in order to generate a preliminary dataset. The preprocessing and filtering may directly correlate each data type to the depth, event, and wellbore of interest while detecting specific changes in certain frequency ranges while filtering out expected incidents. Further, the data filter module 218 may flag the event of interest and quantify the severity of the event based upon the rate of change of the sensor measurement.

The AI model training application 212 may receive the mined and filtered high-quality dataset with one or more machine learning training modules 222 which receive the dataset as an input with a desired output of the specific hazard event of interest. The one or more machine learning training modules 222 may utilize any number of machine learning algorithms including, but not limited to, Logistic Regression (LR), Naïve Beyes (NB), K-Nearest Neighbor (KNN), Decision Tree (DT), Ada Boost (AB), Deep Neural Network (DNN), Support Vector Machine (SVM), or Random Forest (RF). Further, ensemble machine learning techniques may also be used. The one or more machine learning training modules 222 may generate recognition patterns depending on the inputs to the system 200, ranging from the general geological and geographical information of an area to the active readings of one or more sensors 204. In some embodiments, the AI model training application 212, and more specifically the one or more machine learning modules 222, may output a trained AI model which may be utilized by the system 200. In alternate embodiments, the AI model training application 212 may be wrapped into the hazard prediction application 214, such that the one or more machine learning training modules 222 may receive feedback from use of the AI models and further train the AI models.

The hazard prediction application 214 may include a series of modules which receive raw or refined data and utilize the trained AI models to predict hazards prior to or during the drilling of a wellbore. The modules of the hazard prediction application 214 may include a real-time warning module 224, a predictive warning module 226, and a planning module 228. The real-time warning module 224 may involve the receipt of real-time data from one or more sensors 204, and may predict or detect an impending drilling hazard utilizing the training previously received from the AI model training application 212. The real-time warning module 224 may be in communication with one or more pieces of drilling equipment via an internal bus of the system 200, or via the network interface 202, such that the real-time warning module 224 may further control the operation of the drilling equipment. In some embodiments, the real-time warning module 224 alter drilling operations, or take remedial actions upon detecting an impending drilling hazard. In alternate embodiments, the real-time warning module 224 may output a warning to an operator or drilling crew of the impending drilling hazard event.

Similarly, the predictive warning module 226 may receive historical drilling data, geographical data, geological data, and real-time sensor data from one or more sensors 204. The predictive warning module 226 may be utilized during the construction of wells at any point from initial drilling to the completion of the well to predict upcoming hazard events. The predictive warning module 226 may utilize the variety of input data to detect upcoming hazard events and warn an operator or drilling crew. In some embodiments, the predictive warning module 226 may provide possible adjustments to the drilling operations to avoid the upcoming hazard events along with a detailed report of the upcoming drilling hazard event. For wells in the planning stage, the planning module 228 may receive geological and geographical data, or 3D models, of an area and a subterranean formation and may suggest one or more possible locations and trajectory for drilling which may be optimized to avoid drilling hazard events. The planning module 228 may provide a heat-map of ideal and dangerous locations for drilling in an area, as well as possible outputs of a drilled well or the formation as a whole, based upon the training data utilized in the training of the AI model. The model may also provide recommended drilling parameters during the planning phase to avoid parameter induced hazards.

In view of the foregoing structural and functional description, those skilled in the art will appreciate that portions of the embodiments may be embodied as a method, data processing system, or computer program product. Accordingly, these portions of the present embodiments may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware, such as shown and described with respect to the computer system of FIG. 3. Furthermore, portions of the embodiments may be a computer program product on a computer-usable storage medium having computer readable program code on the medium. Any non-transitory, tangible storage media possessing structure may be utilized including, but not limited to, static and dynamic storage devices, hard disks, optical storage devices, and magnetic storage devices, but excludes any medium that is not eligible for patent protection under 35 U.S.C. § 101 (such as a propagating electrical or electromagnetic signals per se). As an example and not by way of limitation, computer-readable storage media may include a semiconductor-based circuit or device or other IC (such, as for example, a field-programmable gate array (FPGA) or an ASIC), a hard disk, an HDD, a hybrid hard drive (HHD), an optical disc, an optical disc drive (ODD), a magneto-optical disc, a magneto-optical drive, a floppy disk, a floppy disk drive (FDD), magnetic tape, a holographic storage medium, a solid-state drive (SSD), a RAM-drive, a SECURE DIGITAL card, a SECURE DIGITAL drive, or another suitable computer-readable storage medium or a combination of two or more of these, where appropriate. A computer-readable non-transitory storage medium may be volatile, nonvolatile, or a combination of volatile and non-volatile, as appropriate.

Certain embodiments have also been described herein with reference to block illustrations of methods, systems, and computer program products. It will be understood that blocks and/or combinations of blocks in the illustrations, as well as methods or steps or acts or processes described herein, can be implemented by a computer program comprising a routine of set instructions stored in a machine-readable storage medium as described herein. These instructions may be provided to one or more processors of a general purpose computer, special purpose computer, or other programmable data processing apparatus (or a combination of devices and circuits) to produce a machine, such that the instructions of the machine, when executed by the processor, implement the functions specified in the block or blocks, or in the acts, steps, methods and processes described herein.

These processor-executable instructions may also be stored in computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory result in an article of manufacture including instructions which implement the function specified. The computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart block or blocks.

In this regard, FIG. 3 illustrates one example of a computer system 300 that can be employed to execute one or more embodiments of the present disclosure. Computer system 300 can be implemented on one or more general purpose networked computer systems, embedded computer systems, routers, switches, server devices, client devices, various intermediate devices/nodes or standalone computer systems. Additionally, computer system 300 can be implemented on various mobile clients such as, for example, a personal digital assistant (PDA), laptop computer, pager, and the like, provided it includes sufficient processing capabilities.

Computer system 300 includes processing unit 302, system memory 304, and system bus 306 that couples various system components, including the system memory 304, to processing unit 302. Dual microprocessors and other multi-processor architectures also can be used as processing unit 302. The system memory 304 may further include an artificial intelligence inference engine 305, which may run one or more artificial intelligence models. System bus 306 may be any of several types of bus structure including a memory bus or memory controller, a peripheral bus, and a local bus using any of a variety of bus architectures. System memory 304 includes read only memory (ROM) 310 and random access memory (RAM) 312. A basic input/output system (BIOS) 314 can reside in ROM 310 containing the basic routines that help to transfer information among elements within computer system 300.

Computer system 300 can include a hard disk drive 316, magnetic disk drive 318, e.g., to read from or write to removable disk, 320, and an optical disk drive 322, e.g., for reading CD-ROM disk 324 or to read from or write to other optical media. Hard disk drive 316, magnetic disk drive 318, and optical disk drive 322 are connected to system bus 306 by a hard disk drive interface 326, a magnetic disk drive interface 328, and an optical drive interface 330, respectively. The drives and associated computer-readable media provide nonvolatile storage of data, data structures, and computer-executable instructions for computer system 300. Although the description of computer-readable media above refers to a hard disk, a removable magnetic disk and a CD, other types of media that are readable by a computer, such as magnetic cassettes, flash memory cards, digital video disks and the like, in a variety of forms, may also be used in the operating environment; further, any such media may contain computer-executable instructions for implementing one or more parts of embodiments shown and described herein.

A number of program modules may be stored in drives and RAM 310, including operating system 332, one or more application programs 334, other program modules 336, and program data 338. In some examples, the application programs 334 can include the data processing application 210, the AI model training application 212, and the hazard prediction application 214, and the program data 338 can include a high-quality dataset, a trained AI model, or any data stored on the database 206. The application programs 334 and program data 338 can include functions and methods programmed to receive and transform data to train an artificial intelligence to predict and prevent drilling hazard events, such as shown and described herein.

A user may enter commands and information into computer system 300 through one or more input devices 340, such as a pointing device (e.g., a mouse, touch screen), keyboard, microphone, joystick, game pad, scanner, and the like. These and other input devices 340 are often connected to processing unit 302 through a corresponding port interface 342 that is coupled to the system bus, but may be connected by other interfaces, such as a parallel port, serial port, or universal serial bus (USB). One or more output devices 344 (e.g., display, a monitor, printer, projector, or other type of displaying device) is also connected to system bus 306 via interface 346, such as a video adapter.

Computer system 300 may operate in a networked environment using logical connections to one or more remote computers, such as remote computer 348. Remote computer 348 may be a workstation, computer system, router, peer device, or other common network node, and typically includes many or all the elements described relative to computer system 300. The logical connections, schematically indicated at 350, can include a local area network (LAN) and/or a wide area network (WAN), or a combination of these, and can be in a cloud-type architecture, for example configured as private clouds, public clouds, hybrid clouds, and multi-clouds. When used in a LAN networking environment, computer system 300 can be connected to the local network through a network interface or adapter 352. When used in a WAN networking environment, computer system 300 can include a modem, or can be connected to a communications server on the LAN. The modem, which may be internal or external, can be connected to system bus 306 via an appropriate port interface. In a networked environment, application programs 334 or program data 338 depicted relative to computer system 300, or portions thereof, may be stored in a remote memory storage device 354.

The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. As used herein, for example, the singular forms “a,” “an,” and “the” are intended to include the plural forms as well, unless the context clearly indicates otherwise. It will be further understood that the terms “contains”, “containing”, “includes”, “including,” “comprises”, and/or “comprising,” and variations thereof, when used in this specification, specify the presence of stated features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof.

Terms of orientation used herein are merely for purposes of convention and referencing and are not to be construed as limiting. However, it is recognized these terms could be used with reference to an operator or user. Accordingly, no limitations are implied or to be inferred. In addition, the use of ordinal numbers (e.g., first, second, third, etc.) is for distinction and not counting. For example, the use of “third” does not imply there must be a corresponding “first” or “second.” Also, if used herein, the terms “coupled” or “coupled to” or “connected” or “connected to” or “attached” or “attached to” may indicate establishing either a direct or indirect connection, and is not limited to either unless expressly referenced as such.

While the disclosure has described several exemplary embodiments, it will be understood by those skilled in the art that various changes can be made, and equivalents can be substituted for elements thereof, without departing from the spirit and scope of the invention. In addition, many modifications will be appreciated by those skilled in the art to adapt a particular instrument, situation, or material to embodiments of the disclosure without departing from the essential scope thereof. Therefore, it is intended that the invention not be limited to the particular embodiments disclosed, or to the best mode contemplated for carrying out this invention, but that the invention will include all embodiments falling within the scope of the appended claims. Moreover, reference in the appended claims to an apparatus or system or a component of an apparatus or system being adapted to, arranged to, capable of, configured to, enabled to, operable to, or operative to perform a particular function encompasses that apparatus, system, or component, whether or not it or that particular function is activated, turned on, or unlocked, as long as that apparatus, system, or component is so adapted, arranged, capable, configured, enabled, operable, or operative.

Claims

1. A method comprising:

inputting a dataset comprising data about a wellbore, a geographical area, a hydrocarbon formation, or any combination thereof, to one or more artificial intelligence models;

generating a prediction, via the one or more artificial intelligence models, of a probability of a hazard event for the wellbore, an impending hazard event during active drilling of the wellbore, an optimal location for the wellbore in the geographical area, or any combination thereof; and

performing or modifying drilling operations in response to the prediction generated by the one or more artificial intelligence models.

2. The method of claim 1, wherein the data comprises a drilling report containing hazard event information, static data about the wellbore, sensor data, or any combination thereof.

3. The method of claim 1, further comprising:

collecting data about one or more existing wellbores, one or more drilled geographical areas, one or more drilled hydrocarbon formation, or any combination thereof;

mining, sorting, and filtering the data to correlate the data to a previous hazard event within the one or more existing wellbores;

flagging the previous hazard event and quantifying the severity of the previous hazard event;

generating a training dataset comprising data sorted by depth, time, and event; and

training the one or more artificial intelligence models on the training dataset and the previous hazard event.

4. The method of claim 3, further comprising:

training the one or more artificial intelligence models on feedback comprising the prediction generated by the one or more artificial intelligence models and an outcome of the performed or modified drilling operations.

5. The method of claim 3, wherein mining the data comprises using natural language processing algorithms for extracting contextual information.

6. The method of claim 3, wherein the one or more artificial intelligence models are trained using one or more machine learning techniques selected from the group consisting of logistic regression, naïve Beyes, k-nearest neighbor, decision tree, Ada boost, deep neural network, random forest, and any combination thereof.

7. The method of claim 3, wherein filtering the data comprises identifying a change in data within a predefined range and excluding data correlated to expected drilling events.

8. The method of claim 1, wherein the drilling operations are performed or modified by the one or more artificial intelligence models in response to the prediction generated by the one or more artificial intelligence models.

9. The method of claim 1, further comprising generating a warning alert to an operator or a drilling crew about the probability of the hazard event or the impending hazard event.

10. A system comprising:

a data processing application comprising a data mining module and a data filter module for receiving, extracting, and filtering a dataset;

an artificial intelligence model training application comprising one or more machine learning training modules; and

a hazard prediction application configured to predict a hazard event within a wellbore utilizing one or more artificial intelligence models.

11. The system of claim 10, wherein the hazard prediction application comprises a real-time warning module configured to predict the hazard event during active drilling, a predictive warning module configured to determine a probability of a future hazard event in the wellbore, a planning module configured to predict an optimal well location in a geographical area, or any combination thereof.

12. The system of claim 10, further comprising one or more sensors within the wellbore configured to provide real-time readings to the system.

13. The system of claim 10, further comprising a database storing historical data comprising hazard event reports, static wellbore data, sensor readings, or any combination thereof from existing wellbores.

14. The system of claim 13, wherein the historical data stored on the database is input to the artificial intelligence model training application for training or updating the one or more artificial intelligence models.

15. The system of claim 10, further comprising one or more pieces of drilling equipment, wherein operations of the one or more pieces of drilling equipment are controlled or modified by the hazard prediction application upon predicting the hazard event.