Patent Grant
11946366
Drill cuttings are an important source of information that is directly available at a well site. Mineralogy and lithology properties of the formation being drilled can be determined through laboratory measurements of the drill cuttings. During drilling, the drilling fluid constantly circulates and enters a shaker, bringing with it pieces of the formation. Further, laboratory measurements make it possible to determine the composition and physical and chemical properties of the formation that is currently being drilled. Upon knowing these formation properties, geologists and engineers can make effective decisions on hydrocarbon drilling and production, and further accurately pick casing points, formation tops, and perforation zones. Current procedures for formation properties determinations are heavily dependent on time-consuming laboratory measurements and a geologist's experience, and thus, may involve time delays and be subject to human error.
This summary is provided to introduce a selection of concepts that are further described below in the detailed description. This summary is not intended to identify key or essential features of the claimed subject matter, nor is it intended to be used as an aid in limiting the scope of the claimed subject matter.
In one aspect, embodiments disclosed herein relate to a method for formation properties prediction in near-real time. The method includes obtaining, by a computer processor, lab measurements of existing drill cuttings at a plurality of depths of a first well. The method includes obtaining, by the computer processor, historical drilling surface data at the plurality of depths from a plurality of wells. The method includes obtaining, by the computer processor, real-time digital photos and real-time drilling surface data of new drill cuttings at a new depth of a new well. The method includes generating, by the computer processor using a prediction model, predicted formation properties of the new drill cuttings based on the real-time digital photos, the real-time drilling surface data, and the new depth. The method further includes predicting, by the computer processor using a near-real-time model and the predicted formation properties, near-real-time formation properties in the new well, wherein the prediction model comprises a historical model that correlates the lab measurements of the existing drill cuttings, and the historical drilling surface data from the plurality of wells, by employing a machine-learning and deep learning algorithms.
According to one aspect, embodiments disclosed herein relate to a system for formation properties prediction in near-real time. The system includes a plurality of formation properties data and a formation properties manager comprising a computer processor. The formation properties manager obtains lab measurements of existing drill cuttings at a plurality of depths of a first well. The formation properties manager obtains historical drilling surface data at the plurality of depths from a plurality of wells. The formation properties manager obtains real-time digital photos and real-time drilling surface data of new drill cuttings at a new depth of a new well. The formation properties manager generates, using a prediction model, predicted formation properties of the new drill cuttings based on the real-time digital photos, the real-time drilling surface data, and the new depth. The formation properties manager further predicts, using a near-real-time model and the predicted formation properties, near-real-time formation properties in the new well, wherein the prediction model comprises a historical model that correlates the lab measurements of the existing drill cuttings, and the historical drilling surface data from the plurality wells, by employing a machine-learning algorithm.
According to one aspect, embodiments disclosed herein relate to s non-transitory computer readable medium storing instructions. The instructions obtain lab measurements of existing drill cuttings at a plurality of depths of a first well. The instructions obtain historical drilling surface data at the plurality of depths from a plurality of wells. The instructions obtain real-time digital photos and real-time drilling surface data of new drill cuttings at a new depth of a new well. The instructions generate, using a prediction model, predicted formation properties of the new drill cuttings based on the real-time digital photos, the real-time drilling surface data, and the new depth. The instructions further predict, using a near-real-time model and the predicted formation properties, near-real-time formation properties in the new well, wherein the prediction model comprises a historical model that correlates the lab measurements of the existing drill cuttings and the historical drilling surface data from the plurality of wells, by employing a machine-learning algorithm.
Other aspects and advantages of the claimed subject matter will be apparent from the following description and the appended claims.
Specific embodiments of the disclosed technology will now be described in detail with reference to the accompanying figures. Like elements in the various figures are denoted by like reference numerals for consistency.
Specific embodiments of the disclosure will now be described in detail with reference to the accompanying figures. Like elements in the various figures are denoted by like reference numerals for consistency.
In the following detailed description of embodiments of the disclosure, numerous specific details are set forth in order to provide a more thorough understanding of the disclosure. However, it will be apparent to one of ordinary skill in the art that the disclosure 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.
Throughout the application, ordinal numbers (e.g., first, second, third, etc.) may be used as an adjective for an element (i.e., any noun in the application). The use of ordinal numbers is not to imply or create any particular ordering of the elements nor to limit any element to being only a single element unless expressly disclosed, such as using the terms “before”, “after”, “single”, and other such terminology. Rather, the use of ordinal numbers is to distinguish between the elements. By way of an example, a first element is distinct from a second element, and the first element may encompass more than one element and succeed (or precede) the second element in an ordering of elements.
In general, embodiments of the disclosure include a system and a method for formation properties prediction in near-real time. More specifically, the present disclosure relates to methods for automated analysis of drill cuttings received at the surface from a well bore, analyzing drilling surface data, utilizing historical drilling and laboratory data, and predicting formation in near real-time by using drill cuttings images. In some embodiments, the method may utilize training data from existing wells to generate a historical model. Further, the method may utilize a prediction model including outputs of the historical model and real-time data from a new well to generate predicted formation properties for the new well.
Furthermore, the method may utilize a near-real-time model and the predicted formation properties to predict near-real-time formation properties ahead of the drill bit in the new well. In some embodiments, the historical model may utilize machine learning (ML) algorithms. Accordingly, timely analysis and prediction of the formation properties of the new well is achieved, human errors are avoided and/or reduced, and historical data and behaviors may be fully utilized.
Turning to
In one or more embodiments, the training data (220) may include lab measurements (e.g., lab measurements (221)) and historical data (e.g., historical data (222)). Detailed contents of the lab measurements (221) and the historical data (222) will be further explained below.
Specifically, the lab measurements (221) may refer to mineralogy data, lithology data, and digital photos of existing drill cuttings collected from at least one of the training wells (111, 112, 113, 114) at various depths. In some embodiments, drill cuttings may refer to broken bits of solid material removed from a drilled borehole. The drill cuttings are carried to the surface of a well by circulating up drilling fluid, and can be separated from the drilling fluid by shale shakers. Mineralogy data specifies scientific study related to a mineral, including chemistry properties, crystal structure, and physical properties. Lithology data specifies physical characteristics of a rock, including color, texture, grains size, grain shape, and composition. The digital photos of the existing drill cuttings may be images captured and produced by cameras containing arrays of electronic photodetectors. The digital photos are digitalized images and are stored as computer files ready for further digital processing and viewing.
Further, the historical data (222) may refer to drilling surface data collected from at least one of the training wells (111, 112, 113, 114) at the various depths. In particular, in some embodiments, the drilling surface data may include rate of penetration (ROP), weight on bit (WOB), SPP (standpipe pressure), logging-while-drilling (LWD), and hookload.
More specifically, the ROP refers to the speed at which a drill bit breaks the rock under it to deepen a borehole. While drilling, the ROP increases in fast drilling formations and decreases in slow drilling formations. The ROP can be expressed as either distance drilled per unit of time (e.g., feet per hour) or time per distance drilled (e.g., minutes per foot). The WOB refers to the amount of downward force exerted on a drill bit during drilling operations. The WOB is usually measured in thousands of pounds and is provided by thick-walled drilled collars. The WOB provides force for the drill bit in order to effectively break the rock.
Continuing with the historical data (222), the SPP refers to the total pressure loss in a system that occurs due to fluid friction. The SPP is a summation of pressure loss in annulus, pressure loss in drill string, pressure loss in bottom hole assembly (BHA), and pressure loss across the bit. The SPP is highly related to jet bit nozzle size selection and flow rate of the cleaning fluid determination, in order to ensure efficient cleaning of the drilled borehole and proper selection of mud pump liner. The LWD refers to measurement of formation properties during the excavation of or shortly after the borehole, through tools integrated into the BHA. The LWD has the advantage of measuring properties of a formation before drilling fluids invade deeply, and timely LWD data can be used to guide well placement, particularly in the zone of interests or in the most productive portion of the formation reservoir. Hookload refers to the actual weight of the drill string measured from the surface. Knowing the hookload helps a drilling person to control weight on bit and decide to increase or decrease the weight imposed on the drill bit.
In some embodiments, the real-time data (230) may include new well data (e.g., new well data (231)). The new well data (231) may refer to real-time drilling surface data and real-time digital photos of new drill cuttings collected from the new well (121) at one or more new/different depths, as well as the actual depth at the time when these data are collected.
The drilling surface data of the new drill cuttings from the new well may also include real-time collected ROP, WOB, SPP, LWD, and hookload as described above.
Keeping with
Continuing with the data controller (250), in addition to collecting and processing the training data (220), the real-time data (230) in different formats may be collected and processed in a similar fashion by the data controller (250) and the data processors (251, 252, 253).
In one or more embodiments, the data controller (250) may be coupled with the formation properties manager (260). In some embodiments, the formation properties manager (260) may be software and/or hardware implemented on the same or a different computing device as the data controller, and may include functionality for detecting and/or managing formation properties. For example, the formation properties manager (260) may collect processed training data (e.g., processed training data (255)) and processed real-time data (e.g., processed real-time data (256)) from the coupled data controller (250). Further, the formation properties manager (260) may include functionality to generate a historical model (e.g., historical model (280)) by utilizing the processed training data (255) from the data controller (250) and applying a machine-learning algorithm that will be explained below.
In one or more embodiments, the formation properties manager (260) may include a prediction model (e.g., prediction model (270)) that generates predicted formation properties (e.g., predicted formation properties (275)) of the new well based on the collected real-time data (230) of the new well. Moreover, the formation properties manager (260) may include a near-real-time model (e.g., near-real-time model (290)). The near-real-time model (290) may be one or more trained machine learning model that includes functionality to predict formation properties in near-real-time (e.g., near-real-time formation properties prediction (295)) ahead of the drill bit.
In some embodiments, the formation properties data source (210), the data controller (250), and the formation properties manger (260) may be implemented on the same computing device, or different computing systems connected by a network. In some embodiments, the formation properties data source (210), the data controller (250), the formation properties manager (260), and/or other elements, including but not limited to network elements, user equipment, user devices, servers, and/or network storage devices may be implemented on computing systems similar to the computing system (500) shown and described in
Continuing with
Turning to
Machine learning models include supervised machine learning models and unsupervised machine learning models. More specifically, supervised machine learning models include classification, regression models, etc. Unsupervised machine learning models include, for example, clustering models. Deep-learning algorithms are a part of machine learning methods based on artificial neural networks with representation learning. For example, a deep-learning algorithm may run data through multiple layers of neural network algorithms, each of which passes a simplified representation of the data to the next layer. More specifically, each artificial neural network consists a plurality of neurons that are staked next to each other and organized in layers. Each neuron may receive various inputs, multiplies the inputs by weights, sums them up, and applies a non-linear function. Deep-learning algorithms are particularly used when a large number of parameters are involved and require access to a vast amount of data to be effective, for example, images process involving millions of features. In one or more embodiments, the deep-learning algorithm (385) may utilize one or more neural network architectures, such as but not limited to, convolutional neural networks, recurrent neural networks, general adversarial neural networks, deep belief networks, autoencoders, etc.
Further, a prediction model (e.g., prediction model (370)) that utilizes the output of the historical model (380) obtains a plurality of processed real-time data (e.g., processed real-time data (330)) of a new well as inputs. In particular, the processed real-time data (330) may include data representing real-time digital photos (e.g., real-time digital photos data (331)), data representing real-time drilling surface data (e.g., real-time drilling surface data (332)), and data representing new depth (e.g., new depth data (333)) at where the aforementioned data are collected. Based on these inputs and the historical model (380), the prediction model (370) outputs predicted lithology data (e.g., predicted lithology data (376)) and predicted mineralogy data (e.g., predicted mineralogy data (377)) in real-time in the borehole being drilled, and predicted ROP (e.g., predicted ROP (378)) of the drill bit in real-time. Predicted lithology data (376) may include formation grain size and shape, as well as mineralogy content, color, and oil shows.
Keeping with
In addition, similar to the trained historical model (380), the near-real-time model (390) may be one or more machine learning models that employ the deep-learning algorithms as described above.
While
Turning to
In Block 410, lab measurements of existing drill cuttings are obtained. For example, lab measurements including lithology data, minerology data, and digital photos of existing drill cutting are collected from a plurality of depths of a training well. The lab measurements may be obtained by a data controller.
In Block 420, historical data of a plurality of training wells are obtained. For example, historical data including drilling surface data at the plurality of depths among the plurality of the training wells. In particular, the drilling surface data may include ROP, WOB, SPP, and LWD at the plurality of depths. The historical data may be obtained by the data controller.
In Block 430, the lab measurements are pre-processed in a single format. For example, the digital photos included in the lab measurements may in various formats, and a data processor comprised in the data controller may process the obtained digital photos and convert them in a single format. The obtained lithology data and minerology data may be processed in a similar manner.
In Block 440, the historical data are pre-processed in a single format. For example, the various drilling surface data may in different formats, and another data processor comprised in the data controller may process the drilling surface data so that file formats of these data are unified. The formats of the lab measurements and the historical data may or may not be the same after the preprocessing occurs in Blocks 430 and 440.
In Block 450, a historical model is generated. In particular, the historical model is generated utilizing the processed lab measurements and the processed historical data, and by employing a deep-learning algorithm, or any other suitable machine learning algorithm. For example, the historical model applies the deep-learning algorithm to correlate the parameters of the lab measurements and the historical data to each other. As a result, the historical model may generate corresponding outputs when new parameters are entered, wherein the new parameters and the corresponding outputs are within the scope of the lab measurements and the historical data.
In Block 460, real-time data of new drill cuttings of a new well are obtained. In particular, the real-time data may include digital photos of the new drill cuttings at a new depth, drilling surface data of the new well at the new depth, and the new depth. The real-time data reflect parameters of the new well at the new depth and at the time when the real-time data are collected.
In Block 470, formation properties of the new well are predicted. For example, the obtained real-time data from Block 460 are entered into a prediction model including the historical model, and the prediction model predicts formation properties of the new well at the new depth and at the time when the real-time data are collected. However, during the procedure of Blocks 460 and 470, the drill bit continuously moves along a borehole. As such, when the predicted formation properties are outputted by the prediction model, the predicted formation properties may be same as or different from the formation properties at the latest location of the drill bit.
In Block 480, near-real-time formation properties are predicted. For example, the predicted formation properties from Block 470 are entered in a near-real-time model that further predicts the near-real-time formation properties of the new well ahead of the drill bit. As a result, the near-real-time formation properties that more accurately reflect the formation properties of the new well at a depth ahead of the drill bit at the current moment are achieved. In particular, the near-real-time formation may be a machine-learning model. The process ends after Block 480
Those skilled in the art will appreciate that the process of
The computer processor(s) (502) may be an integrated circuit for processing instructions. For example, the computer processor(s) may be one or more cores or micro-cores of a processor. The computing system (500) may also include one or more input devices (510), such as a touchscreen, keyboard, mouse, microphone, touchpad, electronic pen, or any other type of input device. In one or more embodiments, the computer processor(s) (502) may be included in the formation properties manager (260) as described in
The communication interface (512) may include an integrated circuit for connecting the computing system (500) to a network (not shown) (e.g., a local area network (LAN), a wide area network (WAN) such as the Internet, mobile network, or any other type of network) and/or to another device, such as another computing device.
Further, the computing system (500) may include one or more output devices (508), such as a screen (e.g., a liquid crystal display (LCD), a plasma display, touchscreen, cathode ray tube (CRT) monitor, projector, or other display device), a printer, external storage, or any other output device. One or more of the output devices may be the same or different from the input device(s). The input and output device(s) may be locally or remotely connected to the computer processor(s) (502), non-persistent storage (504), and persistent storage (506). Many different types of computing systems exist, and the aforementioned input and output device(s) may take other forms. In one or more embodiments, the one or more output devices (508) may be included in the formation properties manager (260) in order to output the near-real-time formation properties prediction (295) as described in
Software instructions in the form of computer readable program code to perform embodiments of the disclosure may be stored, in whole or in part, temporarily or permanently, on a non-transitory computer readable medium such as a CD, DVD, storage device, a diskette, a tape, flash memory, physical memory, or any other computer readable storage medium. Specifically, the software instructions may correspond to computer readable program code that, when executed by a processor(s), is configured to perform one or more embodiments of the disclosure.
The computing system (500) in
The computing system of
For example, a GUI may first obtain a notification from a software application requesting that a particular data object be presented within the GUI. Next, the GUI may determine a data object type associated with the particular data object, e.g., by obtaining data from a data attribute within the data object that identifies the data object type. Then, the GUI may determine any rules designated for displaying that data object type, e.g., rules specified by a software framework for a data object class or according to any local parameters defined by the GUI for presenting that data object type. Finally, the GUI may obtain data values from the particular data object and render a visual representation of the data values within a display device according to the designated rules for that data object type.
Data may also be presented through various audio methods. In particular, data may be rendered into an audio format and presented as sound through one or more speakers operably connected to a computing device.
Data may also be presented to a user through haptic methods. For example, haptic methods may include vibrations or other physical signals generated by the computing system. For example, data may be presented to a user using a vibration generated by a handheld computer device with a predefined duration and intensity of the vibration to communicate the data.
The above description of functions presents only a few examples of functions performed by the computing system of
While the disclosure has been described with respect to a limited number of embodiments, those skilled in the art, having benefit of this disclosure, will appreciate that other embodiments can be devised which do not depart from the scope of the disclosure as disclosed herein. Accordingly, the scope of the disclosure should be limited only by the attached claims.
Although only a few example embodiments have been described in detail above, those skilled in the art will readily appreciate that many modifications are possible in the example embodiments without materially departing from this invention. Accordingly, all such modifications are intended to be included within the scope of this disclosure as defined in the following claims. In the claims, 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. Thus, although a nail and a screw may not be structural equivalents in that a nail employs a cylindrical surface to secure wooden parts together, whereas a screw employs a helical surface, in the environment of fastening wooden parts, a nail and a screw may be equivalent structures. It is the express intention of the applicant not to invoke 35 U.S.C. § 112, paragraph 6 for any limitations of any of the claims herein, except for those in which the claim expressly uses the words ‘means for’ together with an associated function.
| Number | Name | Date | Kind |
|---|---|---|---|
| 7079952 | Thomas et al. | Jul 2006 | B2 |
| 8705318 | Zheng et al. | Apr 2014 | B2 |
| 9016399 | Pelletier et al. | Apr 2015 | B2 |
| 20040256152 | Dashevskiy et al. | Dec 2004 | A1 |
| 20140351183 | Germain | Nov 2014 | A1 |
| 20150241591 | Burmester | Aug 2015 | A1 |
| 20170058620 | Torrione | Mar 2017 | A1 |
| 20190147125 | Yu | May 2019 | A1 |
| 20200040719 | Maniar | Feb 2020 | A1 |
| 20210319257 | Francois | Oct 2021 | A1 |
| 20220107279 | Krumm | Apr 2022 | A1 |
| 20220120168 | Katterbauer | Apr 2022 | A1 |
| Number | Date | Country |
|---|---|---|
| 2794094 | Feb 2020 | CA |
| 2552939 | Feb 2018 | GB |
| 2552939 | Feb 2018 | GB |
| 2008094944 | Aug 2008 | WO |
| 2013089683 | Jun 2013 | WO |
| 2016077521 | May 2016 | WO |
| 2016154723 | Oct 2016 | WO |
| Entry |
|---|
| Gupta et al. “Looking Ahead of the Bit Using Surface Drilling and Petrophysical Data: Machine-Learning-Based Real-Time Geosteering in Volve Field”, 2020 (Year: 2020). |
| International Search Report and Written Opinion issued in Application No. PCT/US2022/015982, dated Apr. 7, 2022 (13 pages). |
| Number | Date | Country | |
|---|---|---|---|
| 20220251951 A1 | Aug 2022 | US |
