System and method for estimating rock particle characteristics based on image and geological information

Abstract

Systems and methods for analyzing rock cuttings and measuring physical lithologic characteristics of rock cuttings are provided. An image analysis workflow is provided that includes a plurality of computing modules to automatically estimate relevant geological information from drilling cuttings. The reference data, the manual description, and the log values are associated and used to determine rock characteristics of the rock cuttings. Software for image analysis is developed and the results are displayed in various views.

Description

[0001] Cross-reference paragraphs

[0002] This application claims the benefit of U.S. Provisional Application No. 63 / 514,691, filed on July 20, 2023, entitled “SYSTEM AND METHOD FOR ESTIMATING ROCK PARTICLE PROPERTIES BASED ON IMAGES AND GEOLOGICAL INFORMATION,” the disclosure of which is incorporated herein by reference. Technical Field

[0003] The present disclosure relates generally to systems and methods for estimating rock grain properties based on image similarity and geological information, and more particularly to analysis of individual rock grains identified in images of the rock grains. Background Art

[0004] This section is intended to introduce the reader to various aspects of the technology that may be related to various aspects of the technology described and / or claimed below. This discussion is believed to be helpful in providing the reader with background information to facilitate a better understanding of the various aspects of the present disclosure. Therefore, it should be understood that these statements should be read in this light and not as any form of admission.

[0005] Rock particles are often the basic unit of post-processing analysis in a specific field and have been widely used in industrial and scientific applications, including but not limited to space exploration, mining, civil engineering, geothermal, and oil and gas. Image data of rock particles usually come from imaging systems that produce digital images or three-dimensional (3D) images from laser scanners. Once rock particles are detected and segmented, they can be used to calculate rock particle properties, such as size, shape, texture and other morphology, rock properties, and rock physical properties to answer questions in specific fields.

[0006] In oil and gas, geothermal, and scientific exploration applications, rock particles are generated during drilling activities. The rock particles are called rock cuttings (or cavings, depending on their size). Rock cuttings are often used to identify lithology types for subsurface characterization and are one of the most available and lowest cost data sources for understanding and characterizing subsurface rock properties. Therefore, there is a strong industry need to automatically analyze rock cuttings to reduce labor costs, improve the accuracy and efficiency of the analysis process, and shorten the turnaround time of interpretation. Summary of the invention

[0007] An overview of certain embodiments described herein is set forth below. It should be understood that these aspects are presented merely to provide the reader with a brief overview of these particular embodiments, and these aspects are not intended to limit the scope of the present disclosure.

[0008] Certain embodiments of the present disclosure include a method, which includes receiving a drill cutting instance image of drill cuttings of a geological formation via an analysis and control system; extracting a set of drill cutting features from the drill cutting instance image via the analysis and control system, and the set of drill cutting features corresponds to a feature vector; querying an index data set using the feature vector via the analysis and control system, and the index data set includes one or more reference feature vectors; identifying multiple reference feature vectors from one or more reference feature vectors via the analysis and control system, and each of the multiple reference feature vectors is associated with a corresponding reference drill cutting; determining a set of drill cutting characteristics of the drill cuttings based on the reference drill cuttings associated with the multiple reference feature vectors via the analysis and control system; and determining a lithological classification of the drill cuttings based on the set of drill cutting characteristics via the analysis and control system.

[0009] Certain embodiments of the present disclosure also include a method, which includes: extracting a corresponding set of reference drill cuttings features from each of one or more reference drill cuttings instance images of a geological formation via an analysis and control system, and each of the one or more reference drill cuttings instance images is associated with a corresponding reference drill cutting, and the corresponding set of reference drill cuttings features corresponds to a corresponding reference feature vector; determining a corresponding set of reference drill cuttings features for the corresponding reference drill cuttings via the analysis and control system, at least based on a guided feature description; and indexing each of the one or more reference drill cuttings instance images with a corresponding set of reference drill cuttings features and a corresponding reference feature vector via the analysis and control system to generate an indexed data set.

[0010] Certain embodiments of the present disclosure also include a system including a drilling device configured to obtain a rock sample from a well, an image acquisition system for acquiring an image of the rock sample, and an analysis and control system. The analysis and control system is configured to obtain a drill cuttings example image for drill cuttings from the image of the rock sample, extract a set of features from the drill cuttings example image based on reference data, and determine a lithology classification of the drill cuttings.

[0011] With respect to various aspects of the present disclosure, various improvements may be made to the above-mentioned features. More features may also be incorporated into these different aspects. These improvements and additional features may exist alone or in any combination. For example, the various features associated with one or more of the illustrated embodiments discussed below may be incorporated into any of the above-mentioned aspects of the present disclosure alone or in any combination. The brief overview given above is intended to familiarize the reader with certain aspects and backgrounds of the embodiments of the present disclosure without limiting the subject matter claimed. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] Various aspects of the present disclosure may be better understood by reading the following detailed description and referring to the accompanying drawings, in which:

[0013] Figure 1 A drilling system according to an embodiment of the present disclosure is shown;

[0014] Figure 2 A shale shaker for removing drill bit cuttings from drilling fluid according to an embodiment of the present disclosure is shown;

[0015] Figure 3 A drill bit that produces drill cuttings according to an embodiment of the present disclosure is shown;

[0016] Figure 4 A system according to an embodiment of the present disclosure is shown, the system comprising a Figure 1 Analysis and control system of drilling system;

[0017] Figure 5 It is shown that according to an embodiment of the present disclosure, Figure 4 The cuttings collection and image acquisition workflow is performed by the analysis and control system;

[0018] Figure 6 An example workflow of lithological characterization of drill cuttings based on image analysis of drill cuttings according to an embodiment of the present disclosure is shown;

[0019] Figure 7 shows a rock sample image and a corresponding segmented individual drill cuttings example image according to an embodiment of the present disclosure;

[0020] Figure 8 The embodiment according to the present disclosure is shown Figure 7 Other embodiments of segmented individual drill cuttings instance images;

[0021] Fig. 9 The embodiment according to the present disclosure is shown Figure 6 An example workflow of the feature estimation process;

[0022] Fig.10 shows a feature vector space according to an embodiment of the present disclosure;

[0023] Fig.11 shows examples of lithology types and geological layers according to embodiments of the present disclosure;

[0024] Fig.12 The embodiment according to the present disclosure is shown in Fig. 9 The user interface used in the workflow;

[0025] Fig.13an output display showing estimated properties of a rock sample image according to an embodiment of the present disclosure; and

[0026] Fig.14 A workflow for property propagation / association according to an embodiment of the present disclosure is shown. DETAILED DESCRIPTION

[0027] One or more specific embodiments of the present disclosure will be described below. These described embodiments are merely examples of the currently disclosed technology. In addition, in order to provide a concise description of these embodiments, all features of the actual implementation may not be described in the specification. It should be understood that in the development of any such actual implementation, as in any engineering or design project, many implementation-specific decisions must be made to achieve the developer's specific goals, such as meeting system-related and business-related constraints, which may vary from implementation to implementation. In addition, it should be understood that such development efforts may be complex and time-consuming, but for ordinary technicians who benefit from the present disclosure, this is still a routine task of design, production and manufacturing.

[0028] When introducing elements of various embodiments of the present disclosure, the articles "a," "an," and "the" are intended to indicate that there are one or more elements. The terms "comprising," "including," and "having" are intended to be inclusive, and mean that there may be additional elements in addition to the listed elements. Furthermore, it should be understood that reference to "one embodiment" or "an embodiment" of the present disclosure is not intended to be interpreted as excluding the existence of additional embodiments that also incorporate the recited features.

[0029] As used herein, the terms “connecting…,” “connecting,” “connected,” “connected with…,” and “connected…” are used to mean “directly connected with…” or “connected with…”; the term “group” is used to mean “one element” or “more than one elements.” In addition, the terms “connecting…,” “connecting…,” “coupled,” “coupled together,” and “connected with…” are used to mean “directly connected together” or “connected together via one or more elements.”

[0030] In addition, as used herein, the terms "real-time," "real-time," or "substantially real-time" may be used interchangeably and are intended to describe operations (e.g., computing operations) that are performed without any human-perceptible interruptions between operations. For example, as used herein, data associated with the systems described herein may be collected, transmitted, and / or used to control computing in "substantially real-time," such that during system operation (e.g., while the system is operating), data reading, data transmission, and / or data processing steps occur once per second, once every 0.1 seconds, once every 0.01 seconds, or even more frequently. In addition, as used herein, the terms "continuous," "continuously," or "continuously" are intended to describe operations that are performed without any noticeable interruptions. For example, as used herein, control commands may be transmitted to certain devices every five minutes, every minute, every 30 seconds, every 15 seconds, every 10 seconds, every 5 seconds, or even more frequently, such that operating parameters of the devices may be adjusted without causing any significant interruption to the closed-loop control of the devices. Furthermore, as used herein, the terms "automatic," "automated," "autonomous," and the like are intended to describe, for example, operations that are performed or caused to be performed by a computing system (i.e., performed solely by the computing system without human intervention). Indeed, although certain operations described herein may not be explicitly described as being performed continuously and / or automatically in substantially real time during operation of a computing system and / or a device controlled by the computing system, it should be understood that such operations may in fact be performed continuously and / or automatically in substantially real time during operation of the computing system and / or a device controlled by the computing system to improve the functionality of the computing system (e.g., by not requiring human intervention, thereby facilitating faster operational decision making, and by, for example, eliminating potential human errors to improve the accuracy of operational decision making), as described in greater detail herein.

[0031] As described above, whenever a drilling process is involved in an activity, rock cuttings are generated and are typically available at the well site. Cuttings properties may include observed or measured geological properties associated with the cuttings. Currently, geoscientists and reservoir engineers in the oil and gas industry generally do not fully utilize cuttings for subsurface characterization. When these rock cuttings are observed and interpreted by human eyes, it is very interpreter-dependent and relatively time-consuming and physically labor-intensive. Therefore, there is a need to automate the cuttings analysis process in the industry. To this end, the embodiments described herein provide a domain-based image analysis workflow that includes multiple computing modules to automatically extract relevant geological information from rock cuttings. As used herein, the terms "image", "digital image", "photo" and "picture" are intended to be used interchangeably. In addition, although described herein as a system and method for analyzing drill bit cuttings images, it should be understood that the embodiments described herein are capable of analyzing images of other types of rock particles, such as other types of cuttings, caves, etc., as well as non-rock objects in mud, such as mud additives, metal shavings, and foreign objects.

[0032] Figure 1 1 shows a drilling system 10 according to an embodiment described herein. As shown, in some embodiments, a drill string 12 may be suspended at an upper end by a kelly and a traveling block 14 and terminated at a lower end by a drill bit 16 (e.g., Figure 3 1). The drill string 12 and drill bit 16 are rotated by a rotary table 18 on a drill floor 20 to drill a borehole 22 in a formation 24, wherein a portion of the borehole 22 may be cased by a casing 26. As shown, in some embodiments, a drilling fluid or drilling "mud" 28 may be pumped by a mud pump 30 to the upper end of the hollow drill string 12 through a connecting mud line 32. From there, the drilling fluid 28 may be pumped downward through the drill string 12, exit the drill string 12 through an opening in the drill bit 16, and return to the surface through an annular space formed between the wall of the borehole 22 and the outer diameter of the drill string 12. Once at the surface, the drilling fluid 28 may be returned via a return flow line 34, such as via a bell shaped joint 36. As shown, in some embodiments, a blowout preventer 38 may be used to prevent a blowout from occurring in the drilling system 10.

[0033] like Figure 1 As shown, drill bit cuttings formed by the drill bit 16 crushing rock in the formation 24 can generally be removed from the returning drilling fluid 28 by a shale shaker 40 in the return flow line 34, so that the drilling fluid 28 can be reused for injection, wherein the shale shaker 40 includes a shaker pit 42 and a gas trap 44. The drilling fluid 28 can then be transported to a mud pit 48, and the mud pump 30 can extract the drilling fluid 28 from the mud pit 48. Figure 2 Drill bit cuttings 46 are shown having been removed from the drilling fluid 28 in the shaker pit 42 of the shale shaker 40. Figure 2 In one embodiment, one or more cameras 52 may be used to capture drill cuttings 46. Figure 3 It is shown how the drill bit 16 generates drill bit cuttings 46 which then flow back upward within the drilling fluid 28 through the annular space formed between the wall of the borehole 22 and the outer diameter of the drill string 12 .

[0034] In addition, if Figure 1 As shown, in some embodiments, an analysis and control system 50 (e.g., a mud logging unit) can be used to control the drilling system 10 and provide analysis of the drill bit cuttings 46, as described in more detail herein. In particular, in some embodiments, the analysis and control system 50 can be configured to automatically analyze the drilling cuttings 46 while performing Figure 1 The drilling system 10 shown is operated by a camera (e.g., Figure 2 The one or more cameras 52 shown in the figure automatically capture images of drill cuttings 46, as described in more detail herein. Figure 2 As shown, in some embodiments, the one or more cameras 52 may be directly associated with the shale shaker 40 (e.g., directly attached to the shale shaker 40 or disposed near or in close proximity to the shale shaker 40). However, in other embodiments, the one or more cameras 52 may be other types of cameras that are not directly associated with the shale shaker 40.

[0035] Figure 4 A system 54 is shown which includes an analysis and control system 50 to monitor and control Figure 1 The drilling system 10 of the present invention may be configured to perform various functions of the embodiments described herein, including but not limited to analyzing images of drill bit cuttings 46 captured by one or more cameras 52 using certain analysis algorithms, as described in more detail herein. In some embodiments, to perform these various functions, the analysis modules 56 execute instructions on one or more processors 58 of the analysis and control system 50, which may be connected to one or more storage media 60 of the analysis and control system 50. In fact, in some embodiments, the one or more analysis modules 56 may be stored in the one or more storage media 60. In some embodiments, the one or more analysis modules 56 include analysis hardware (e.g., analysis circuitry) and / or software including processor executable instructions and related data.

[0036] In some embodiments, the one or more processors 58 may include a microprocessor, a microcontroller, a processor module or subsystem, a programmable integrated circuit, a programmable gate array, a digital signal processor (DSP), or other control or computing device. In some embodiments, the one or more storage media 60 may be implemented as one or more non-transitory computer-readable or machine-readable storage media. In some embodiments, the one or more storage media 60 may include one or more different forms of memory, including semiconductor memory devices, such as dynamic or static random access memory (DRAM or SRAM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), and flash memory; disks, such as fixed disks, floppy disks, and removable disks; other magnetic media including tapes; optical media, such as compact disks (CDs) or digital video disks (DVDs); or other types of storage devices. Note that the processor-executable instructions and associated data of the analysis module 56 may be provided on one computer-readable or machine-readable storage medium of the storage medium 60, or alternatively, may be provided on multiple computer-readable or machine-readable storage media distributed in a large system that may have multiple nodes. Such computer-readable or machine-readable storage media are considered part of an article of manufacture, which may refer to any manufactured component or components. In some embodiments, one or more storage media 60 may be located in a machine running machine-readable instructions, or may be located at a remote site from which machine-readable instructions may be downloaded over a network for execution.

[0037] In certain embodiments, the processor 58 may be connected to a network interface 62 of the analysis and control system 50 to allow the analysis and control system 50 to communicate with various surface sensors 64 (Internet of Things (IoT) sensors, meters, etc.) and / or downhole sensors 66 described herein, as well as actuators 68 and / or PLC 70 of surface equipment 72 and / or downhole equipment 74 for monitoring and / or controlling the operation of the drilling system 10, as described in more detail herein. In addition, in certain embodiments, the network interface 62 may also facilitate the analysis and control system 50 to communicate with one or more cameras 52, as described in more detail herein. In certain embodiments, the network interface 62 may also facilitate the analysis and control system 50 to transmit data to a cloud-based service 76 (or other wired and / or wireless communication network), for example, to archive the data or to enable an external computing system 78 (e.g., in certain embodiments, a cloud-based computing system) to access the data and / or remotely interact with the analysis and control system 50. For example, in certain embodiments, some or all of the analysis modules 56 described in more detail herein may be executed via cloud and edge deployments.

[0038] In certain embodiments, the analysis and control system 50 may include an electronic display 80 configured to display a graphical user interface to present the results of the analysis described herein. In addition, in certain embodiments, the graphical user interface may present other information to the operator of the equipment 72, 74. For example, the graphical user interface may include a dashboard configured to present visual information to the operator. In certain embodiments, the dashboard may display on-site (e.g., real-time) data and the analysis results described herein. In addition, in certain embodiments, the analysis and control system 50 may include one or more input devices 82 configured to enable the operator to, for example, provide commands to the equipment 72, 74 described herein. In addition, in certain embodiments, the display 80 may include a touch screen interface configured to receive input from the operator.

[0039] It should be understood that Figure 4 The system 54 shown in the figure is exemplary only, and the system 54 may have more or fewer components than shown, and may be combined Figure 4 Additional components not shown in the embodiments, and / or the system 54 may have Figure 4 In addition, Figure 4 The various components shown in can be implemented with hardware, software or a combination of hardware and software, including one or more signal processing and / or application-specific integrated circuits. In addition, the operation of the system 54 described herein can be implemented by running one or more functional modules in an information processing device, such as a dedicated chip, such as an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA), a programmable logic device (PLD), a system on a chip (SOC) or other appropriate device. These modules, combinations of these modules and / or their combination with hardware are all included within the scope of the embodiments described herein.

[0040] In conventional systems, drill bit cuttings 46 are analyzed by mud logging tools in a mud logging unit. These conventional systems are often human dependent. The embodiments described herein enhance the analysis of such drill bit cuttings 46. Figure 5 It shows that it can be Figure 4 The analysis and control system 50 performs the drilling cuttings collection and image acquisition workflow 84. Figure 5 As shown, the cuttings collection and image acquisition workflow 84 begins with the drilling system 10 generating rock particles (e.g., including the drill bit cuttings 46 described above). The rock particles are then collected, for example, from the shale shaker 40, as described with respect to FIG. Figure 2 as described (e.g., sample collection 86).

[0041] Then, in some embodiments, the rock particles can be prepared for analysis (e.g., sample preparation 88) by, for example, drying the rock particles in an oven for analysis. In addition, in some embodiments, sample preparation 88 can include screening the rock particles using one or more meshes 90 to select drill cuttings 46 that fall within certain size ranges. In some embodiments, the size of the mesh 90 can range between 0.25 millimeters (mm) and 3.0 mm, and can be about 0.25 mm, about 0.50 mm, about 0.75 mm, about 1.0 mm, about 1.25 mm, about 1.50 mm, about 1.75 mm, about 2.0 mm, about 2.25 mm, about 2.50 mm, about 2.75 mm, or about 3.0 mm. It should be understood that the successive meshes 90 through which the rock particles are screened may begin with larger meshes 90 followed by progressively smaller meshes 90, such that larger drill cuttings 46 are screened earlier and smaller drill cuttings 46 are screened later, until the screened rock particles are so fine that they are no longer considered to be drill cuttings 46 themselves. It should also be understood that the size of the particle drill cuttings 46 is the smallest axis of the drill cuttings 46 when the drill cuttings 46 are approximated as ellipsoids. In certain embodiments, the sample preparation 88 may include placing the screened drill cuttings 46 in a sample tray having a relatively vivid background color (e.g., pure magenta (e.g., RGB values ​​of 255, 0, 255), pure blue (e.g., RGB values ​​of 0, 0, 255), pure green (e.g., RGB values ​​of 0, 255, 0), etc.). Typically, such colors do not exist in nature and, therefore, help the instance segmentation model avoid detecting the background of the sample tray as part of the instance. In some cases, the sample tray may be prepared by a manual or automated system, so the distribution of the drill cuttings 46 is generally random. For example, the drill cuttings 46 may be in contact or accumulated in some areas of the sample tray, while being sparsely distributed in other areas.

[0042] In some embodiments, an image 94 of the drill cuttings 46 may be taken (e.g., image acquisition 96) by an imaging device (e.g., one or more cameras 52). The following examples are given with a camera that detects light in the visible light spectrum, but the same methods can be applied to images taken with infrared (IR) or ultraviolet (UV) cameras that detect light in the UV or IR domain. As described in more detail herein, in some embodiments, during image acquisition 96, the illumination, color, and resolution of the image 94 may be calibrated and standardized to obtain a quantitative and reliable measurement of pixel values ​​between the images 94. For example, in some embodiments, the color / illumination calibration may be obtained by using a colorimetric algorithm for a previously analyzed image 94 and a current image 94 of interest, while the resolution calibration may be based on the lens focal length, focal length, and sensor size / resolution of the current image 94 of interest compared to the previously analyzed image 94. All of these parameters may vary, but the final image is "calibrated" and the same object can be digitized with reasonably close values. Therefore, the pixel value (e.g., the image 94 may be a digital image) and size are considered to be valid measurements of the particles, not just representations. The embodiments described herein are capable of creating such calibrated and error-evaluated input images 94. Without such calibration, the final object classification may vary due to acquisition rather than any real-world differences. In some embodiments, an optical image of a sample may be taken under visible light. In some embodiments, an optical image of the same sample may be taken under UV light, which may cause some minerals and oil to fluoresce (if the oil is trapped in the drill cuttings). This may be a clue to the presence of oil in a certain rock type or geological formation to which the drill cuttings belong. In some embodiments, optical images of the same sample taken under different light spectra (e.g., visible, UV, IR) may be combined for image analysis of the sample, such as Figure 6 and Figure 8 As described in.

[0043] Then, in some embodiments, the image 94 (e.g., a digital image) of the drill cuttings 46 may be compressed to facilitate transmission (e.g., image compression 98). In particular, well sites are often located in relatively remote locations where available network bandwidth may be relatively slow. Therefore, compressing the image 94 of the drill cuttings 46 may facilitate transmission of the image 94. It should be understood that if the well site has a higher bandwidth (e.g., when the well site has wired Internet access), compressing the image 94 of the drill cuttings 46 may not be beneficial. Then, in some embodiments, the image 94 of the drill cuttings 46 may be transmitted (e.g., image transmission 100) to, for example, Figure 4The analysis and control system 50 and / or computing system 78 shown in FIG. 1 are used for analysis, as described in more detail herein. The analysis and control system 50 may be located at the well site or remotely. Then, in certain embodiments, the analysis of the image 94 of the drill cuttings 46 (e.g., image analysis 102) may include extracting geologically meaningful information related to the drill cuttings 46 from the image 94, as described in more detail herein.

[0044] The embodiments described herein determine measurements from an image 94 of drill cuttings 46 related to the lithology of the formation 24 from which the drill cuttings 46 were generated, based on lithologic geoscience definitions. In general, the lithology of a rock unit is a description of its physical characteristics that are visible in outcrops, in hand or core samples, or through a low-power microscope. These physical characteristics include, but are not limited to, color, texture, grain size, and composition. Lithology can be a detailed description of these physical characteristics or a summary of the overall physical characteristics of the rock. Examples of lithologic types can include sandstone, slate, basalt, limestone, and the like. Therefore, color, texture, and grain size are physical characteristics of a type of lithology, and the workflow described herein shows how to measure, extract, and merge these physical characteristics to obtain automatic lithologic image recognition.

[0045] The embodiments described herein are applicable to analyzing the prepared calibration image 94 of drill cuttings 46, regardless of the method used to implement the reference Figure 5 The specific methods of sample preparation 88, image acquisition 96, image compression 98 and image transmission 100 are described.

[0046] In some embodiments, the analysis and control system 50 may be configured to separate pixels of individual rock particles from a plurality of particles (e.g., drill cuttings 46) depicted in an image 94 (e.g., a calibration image). Further, in some embodiments, the analysis and control system 50 is configured to determine relevant morphological (e.g., size, shape, etc.), color, and texture data from one or more individual rock particles (e.g., drill cuttings 46) depicted in the calibration image 94. Further, in some embodiments, the analysis and control system 50 may be configured to utilize information extracted from the calibration image 94 to perform geological / lithological classification at a plurality of different classification levels (e.g., at a single particle / drill cutting level, at a single sample level of a plurality of drill cuttings, at a specific depth interval within the wellbore 22, for a specific geological formation 24, for the entire well, for the entire well site, etc.). As described in more detail herein, incorporating the analysis results at a plurality of classification levels enables an operator to analyze the drill cuttings 46 in a more robust (and automated and more accurate) manner than conventional techniques. Further, in some embodiments, the analysis and control system 50 may be configured to utilize information derived herein based on the calibration image 94 to create a mud logging report. Furthermore, in certain embodiments, the analysis and control system 50 may be configured to output any relevant information from the calibration image 94 that may be integrated with answer products associated with other wells. Furthermore, in certain embodiments, the analysis and control system 50 may be configured to utilize a supervised machine learning model (e.g., another well from the same field or another field with a similar geological setting) to infer the lithology type from the calibration image 94 from the current well. Furthermore, in certain embodiments, the analysis and control system 50 may be configured to utilize all of this type of information to automatically adjust operating parameters of the drilling system 10 (e.g., drilling fluid composition, drill bit speed, etc.), as described in more detail herein.

[0047] Although Figure 5 The drill cuttings collection and image acquisition workflow 84 shown in FIG. 8 has been described with specific reference to analyzing the drill cuttings 46 to determine geologically meaningful information related to the drill cuttings 46 from the images 94 , but the embodiments described herein may be extended to analysis of the images 94 to identify features of other types of objects, not just drill cuttings 46 .

[0048] In certain embodiments, object-based image analysis (OBIA) may be used to analyze the image 94 (e.g., the calibration image). OBIA generally involves grouping multiple pixels in a digital image into shapes that are meaningful representations of objects based on spectral or spatial similarities or external variables (e.g., soil or geological units). Fig. 9 As shown in , OBIA can be used to handle complex classes defined by spatial and hierarchical relationships within and during the classification process. OBIA can include image segmentation, feature extraction, and classification, such as Figure 6As previously described, the calibration image 94 can be used for image analysis to obtain more reliable results. The embodiments described herein focus on feature extraction and classification analysis of the calibration image 94. Certain embodiments implement lithology identification and / or classification of the image 94 from the drill cuttings 46 by applying neural network (NN) and machine learning (ML) techniques.

[0049] Figure 6 An example workflow 110 of the system 54 performing object-based image analysis (OBIA) on an image 94 (e.g., a calibration image) is shown. The image 94 (e.g., a digital image) can be input into a drill cuttings instance segmentation module 112 (e.g., a tool or instrument) of the system 54 to obtain segmented individual drill cuttings instance images 114 (e.g., Figure 7 As shown in Figure 7 and Figure 8 As shown, the drill cuttings instance segmentation module 112 may identify and separate various objects within the calibration image 94 , for example, by detecting the boundaries of each object and assigning each object a unique label. Figure 7 An example of a calibration image 94 and a corresponding segmented individual drill cuttings instance image 114 is shown. In some embodiments, the drill cuttings instance segmentation module 112 may include analysis hardware (e.g., analysis circuitry) and / or software including processor executable instructions and associated data. In some embodiments, the drill cuttings instance segmentation module 112 may utilize a Mask R-CNN (region-based convolutional neural network) to identify various objects in the calibration image 94, which has been found to outperform traditional segmentation methods. The Mask R-CNN model is trained in a supervised manner, and it has been found that the same approach can be used to effectively build a system that runs instance segmentation using any type of image (e.g., digital images, range images, etc.) and rock particle scenes (such as rock cuttings, mining sites, space, etc.) by training the model with relevant data sets.

[0050] The segmented individual drill cuttings instance images 114 may be used in a rock property estimation process 116 and a rock property measurement process 118. For example, in the rock property estimation process 116, the segmented individual drill cuttings instance images 114 may be input into a feature extraction module 120 (e.g., a tool or instrument) of the analysis and control system 50. In certain embodiments, the feature extraction module 120 may include analysis hardware (e.g., analysis circuitry) and / or software including processor-executable instructions and associated data. The feature extraction module 120 may analyze the segmented individual drill cuttings instance images 114 using reference data 122 (e.g., baseline drill cuttings used as a reference) and output a set of parameters (features) representing color and texture characteristics of the drill cutting instances, which may be key visual characteristics for identifying rock types, such as Figure 8 Therefore, the drill cuttings feature is a set of values ​​that represent or summarize the color and texture information of the drill cuttings instance image. Figure 8 An example of segmented individual drill cuttings instance images 114 is shown, where the individual drill cuttings instance images (eg, drill cuttings A and drill cuttings B) are segmented with features corresponding to a number D (eg, feature 1 ,feature 2 …feature D ) is indicated by a set of parameters. Figure 8 As shown, the individual drill cuttings instance images (e.g., drill cuttings A, drill cuttings B) in the segmented individual drill cuttings instance images 114 may have any size, shape (e.g., height (H), width (W)) and color. The feature extraction module 120 may analyze the segmented individual drill cuttings instance images 114, and each drill cuttings instance image (Ii) may be extracted from the image space I i ∈R (Hi×Wi×3) (e.g., height (Hi), width (Wi), 3 color channels (e.g., red, green, blue)) to a D-dimensional feature vector space D-dimensional feature vector space R D It may include D features (e.g., features 1 ,feature 2 …feature D ) axis, and the eigenvector can have along the eigenvector space R D In some embodiments, different types of images representing different physical properties of the same rock particle sample can be obtained without changing the position of the sample in the sample disk. In this case, we can combine these images and generate a single multi-channel image. Therefore, the transformation is from image space I i ∈R (Hi×Wi×C) To D-dimensional feature vector space Where C is the number of channels (e.g., a physical property of a rock particle sample, such as color). For example, in some embodiments, the fluorescence of minerals in drill cuttings may be a clue to certain rock types, and an optical image of the same sample (e.g., drill cuttings 46) that may cause the minerals to fluoresce under ultraviolet light may be combined with calibration image 94 (e.g., obtained using visible light) to create a 6-channel image (e.g., 3 color channels for an image taken with ultraviolet light and 3 color channels for an image taken with visible light). In this case, the transformation would be from image space I to image space C. i ∈R (Hi×Wi×6) To D-dimensional feature vector space A lithology classification module 124 (e.g., a tool or instrument) of the analysis and control system 50 may use the feature vectors of the segmented individual drill cuttings instance images 114 to determine the lithology 126 of the drill cuttings 46, as described in more detail herein. In certain embodiments, the lithology classification module 124 may include analysis hardware (e.g., analysis circuitry) and / or software including processor executable instructions and associated data. For a lithology type of rock (e.g., sandstone), a grain size analysis module 128 (e.g., a tool or instrument) of the analysis and control system 50 may be used to analyze each individual drill cutting 46 to identify a grain size class (e.g., fine, medium, coarse, etc.) 130 and a grain size distribution 132 of each individual drill cutting 46, as well as other grain size-related characteristics. In certain embodiments, the grain size analysis module 128 may include analysis hardware (e.g., analysis circuitry) and / or software including processor executable instructions and associated data.

[0051] The feature vectors of the segmented individual cuttings instance images 114 can be used with other features 134 (e.g., features that are not easily extracted from digital images) to estimate rock properties, as described in more detail herein. For example, the other features 134 can include mannual feature descriptions, associated well log values, or other external data, such as, but not limited to, stratigraphic geological sequences, logging while drilling, 3D model features, physical analysis of the cuttings (e.g., diffraction measurements, calcimetry, acid testing), or any other information that may help resolve ambiguity in lithology classification. In some embodiments, the other features 134 can be determined using instruments / tools (e.g., surface equipment 72, downhole equipment 74), analysis hardware (e.g., analysis circuitry), and / or software including processor executable instructions and associated data.

[0052] like Figure 8As shown, the individual drill cuttings instance images (e.g., drill cuttings A, drill cuttings B) in the segmented individual drill cuttings instance images 114 can have any size, shape, and color. The feature measurement process 118 can use the shape measurement module 136 (e.g., tool or instrument) of the analysis and control system 50, and the segmented individual drill cuttings instance images 114 can be analyzed by the shape measurement module 136 based on the individual instances to determine the size 138 and shape 140 of the individual instances. The feature measurement process 118 can use the texture measurement module 142 (e.g., tool or instrument) of the analysis and control system 50, and the segmented individual drill cuttings instance images 114 can be analyzed by the texture measurement module 142 based on the individual instances to determine the texture 144 of the individual instances, such as homogeneous, heterogeneous, layered, granular, etc. The feature measurement process 118 can also use the color measurement module 146 (e.g., tool or instrument) of the analysis and control system 50, and the segmented individual drill cuttings instance images 114 can be analyzed by the color measurement module 146 based on the individual instances to determine the color 148 of the individual instances. In certain embodiments, the shape measurement module 136, the texture measurement module 142, and the color measurement module 146 may include analysis hardware (e.g., analysis circuitry) and / or software including processor-executable instructions and associated data. The output of the property estimation process 116 (e.g., lithology 126, grain size fraction 130, grain size distribution 132) and the output of the property measurement process 118 may be combined to obtain a combined result 150.

[0053] The feature estimation process 116 may include two stages: stage one, for reference data description and indexing (e.g., baseline drill cuttings for reference); and stage two, for feature estimation of drill cuttings instance images based on the reference data description and indexing obtained in stage one. Fig. 9 A workflow 200 for stage one of data description and indexing using reference data 122 is shown. In some embodiments, the workflow 200 may include instruments / tools (e.g., surface equipment 72, downhole equipment 74), hardware (e.g., analysis circuitry), and / or software including processor executable instructions and associated data. The reference data 122 may include image data of a set of M images (e.g., calibration images 94) obtained when analyzing previous data from a previous well or a current well.

[0054] The workflow 200 may include instance segmentation (process 1), image feature extraction (process 2), feature association (process 3), and indexing (process 4). In process 1 of the workflow 200, each image of the set of M images may be analyzed (e.g., by the drill cuttings instance segmentation module 112) to obtain a plurality of (e.g., N) individual drill cuttings instance images 202 (e.g., I 1 , I 2 ,…IN ) of the corresponding segmented individual drill cuttings instance image 201. The drill cuttings (eg, I 1 ,I 2 ,…I N ) can be used as reference drill cuttings or baseline drill cuttings. In process 2 of workflow 200, the segmented individual drill cuttings instance images can be analyzed (e.g., by feature extraction module 120) to obtain individual drill cuttings instance images 202 (e.g., I 1 ,I 2 ,…I N ) is the number of features of D (e.g., feature 1 ,feature 2 …feature D ) corresponding set of parameters. Therefore, the individual cuttings instance image 202 (e.g., I 1 , I 2 ,…I N ) can be transformed from the image space to the D-dimensional feature vector space R D The corresponding feature vector 204 in (e.g., represented by the corresponding matrix y 1 ,y 2 ,…y N express).

[0055] In process 3 of the workflow 200, a separate drill cuttings instance image 202 (eg, 1 ,I 2 ,…I N ) can be described manually by a geologist (e.g., 3-1 guided manual characterization) guided by drill cuttings characterization software, which is Fig.12 and 13 . The cuttings description software may include a description wizard with multiple entries, such as lithology type, hardness, visual porosity, etc. Some entries of the description wizard may not be easily extracted from the digital image. In process 3 of workflow 200, well logging values ​​can also be used for characteristic description (e.g., 3-2 well logging value association). Well logging can include detailed records of geological formations (e.g., formation 24). The detailed records may include information about geological properties (e.g., lithology, layers, depositional environment) and rock physical characteristics (e.g., water saturation, porosity, permeability, shale volume) of the geological formation, which can be used to control drilling equipment or drilling plans of the drilling system 10. Well logging may be a function of borehole depth. For example, as Fig.14As shown, drill cuttings 46 may be collected at a depth in the borehole 22, and the well log values ​​at that depth may be associated with drill cuttings at the same depth from the same well or a different well. For example, this approach may be used when the reference data in stage one includes a well with drill cuttings and well log measurements, and the drill cuttings instance image in stage two includes another well with drill cuttings but no well log measurements. The corresponding individual drill cuttings instance images 202 (e.g., I may be determined by using guided manual characterization, associated well log values, or other external data, such as, but not limited to, stratigraphic geological sequences, well logging while drilling, 3D model properties, physical analysis of drill cuttings (e.g., diffraction measurements, calcimetry, acid testing), or any other information that may help resolve ambiguity in lithology classification. 1 ,I 2 ,…I N ) characteristics 206 (e.g. P 1 , P 2 ,…P N ). The characteristics 206 may include characteristics of the reference cuttings in the individual cuttings example image 202, such as lithology type, hardness, visual porosity, gamma ray logging, porosity logging, etc. For example, P 1 It may include the description "lithology type: limestone; hardness: 0.6, visual porosity: 1%, gamma log: 30 API; porosity log: 3% ...". 2 It may include the description "lithology type: sandstone; hardness: 0.6, visual porosity: 5%, gamma log: 30API; porosity log: 3%..." ... N It may include a description of “lithology type: sandstone; hardness: 0.7, visual porosity: 2%, gamma ray logging: 30API; porosity logging: 3%…”… Based on feature similarity, the nth reference cuttings (P N ) can be propagated to other drill cuttings in the same segmented individual drill cuttings instance image, and a view 207 (e.g., a two-dimensional (2D) view, a three-dimensional (3D) view) of the drill cuttings characteristic description can be displayed on a user interface.

[0056] In process 4 of workflow 200, a drill cuttings example image (e.g., 1 , I 2 ,…I N ), feature vector 204 (eg, from the corresponding matrix y 1 ,y 2 ,…y N ), and the corresponding features 206 may be indexed and stored together in an index dataset 208 (eg, index A). The index dataset 208 may include drill cuttings example images (eg, I 1 , I 2 ,…I N), the eigenvector 204 (e.g., represented by the corresponding matrix y 1 ,y 2 ,…y N denoted), and the corresponding characteristics 206 of the group M images in the reference data 122 (e.g., P 1 , P 2 , …P N ). In process 4, clustering techniques may be applied to the index data set 208 to reduce the data size of the index. The clustering technique may be a data compression technique or a redundancy reduction technique. For example, when there are multiple drill cuttings instances that are visually very similar and the feature vector space R D When the corresponding feature distances (e.g., squared Euclidean (L2 square) distances) in the image are very close to each other, the drill cuttings instances can be grouped into a cluster, and only representative drill cuttings instances can be retained, where the representative drill cuttings instance can be the centroid drill cuttings instance of the cluster. Multiple index databases can be created by using different groups of M images. For example, various index data sets (e.g., index A, index B, index C) can be created from images of a specific geographic location or a specific geological formation.

[0057] Fig. 9 Also shown is a workflow 210 for estimating properties (e.g., lithology type, hardness, visual porosity, gamma ray log, porosity log) of a drill cuttings instance image using the properties 206 obtained in the workflow 200 and the index data set 208. In some embodiments, the workflow 210 may include instruments / tools (e.g., surface equipment 72, downhole equipment 74), hardware (e.g., analysis circuitry), and / or software including processor-executable instructions and associated data. The workflow 210 may include instance segmentation (process 1), image feature extraction (process 2), querying (process 3), and property propagation (process 4). In process 1 of the workflow 210, an image (e.g., a calibration image 94) may be analyzed (e.g., by a drill cuttings instance segmentation module 112) to obtain segmented individual drill cuttings instance images having multiple (e.g., J) unknown individual drill cuttings instance images 212 (e.g., I 1 ,I 2 ,…I J ) (e.g., unknown lithology type, unknown hardness, unknown visual porosity, unknown gamma ray log, unknown porosity log). In process 2 of workflow 210, the segmented individual cuttings instance images may be analyzed (e.g., by feature extraction module 120) to obtain features for each unknown individual cuttings instance image 212 (e.g., I 1 ,I 2 ,…I J ) D features (e.g., feature 1, feature 2, … feature D). Therefore, each unknown individual cuttings instance image 212 (e.g., I 1 , I 2 ,…I J ) can be transformed from the image space to the D-dimensional feature vector space R D The corresponding eigenvector 214 in (eg, represented by the corresponding matrix x J For a newly acquired image, process 1 and process 2 of the workflow 210 may be repeated.

[0058] Process 3 of workflow 210 may be used to convert feature vector 214 (e.g., by the corresponding matrix x J ) and the feature vector of the index memory (for example, represented by the corresponding matrix y 1 ,y 2 ,…y N ) for comparison, from each unknown individual cuttings instance image 212 (e.g., I 1 ,I 2 ,…I J ) is retrieved from the index memory (e.g., index A, index B, index C) of the number K most similar (K ≥ 1) drill cuttings 216. As described above, the characteristics of the number K most similar drill cuttings 216 (e.g., lithology type, hardness, visual porosity, gamma ray logging, porosity logging) can be obtained in the characteristic association of stage one. The feature vector space R can be calculated D The eigenvectors 214 in (eg, represented by the corresponding matrix x J ) and the feature vectors of the index memory (e.g., represented by the corresponding matrix y 1 ,y 2 ,…y N denoted by the distance (e.g., squared Euclidean (L2 square) distance) between them, and a nearest neighbor search method can be used to determine the number of K most similar drill cuttings 216 by finding the number of K minimum distances, such as Fig.10 shown.

[0059] Fig.10 is a D-dimensional feature vector space R D 220 (for example, having two features V 1 and V 2 The eigenvector space R can be calculated. D The feature vector of the drill cuttings example image 222 in 220 (eg, represented by the corresponding matrix x J ) and the feature vectors of the index memory (e.g., represented by the corresponding matrix y 1 ,y 2 ,…y N The distance d between i(e.g., i=1, 2...N) (e.g., squared Euclidean distance (L2 square)), and a nearest neighbor search method can be used to determine the number K of most similar rock particles (e.g., K=1, 2, 3...). For example, in Fig.10 When K=3, the feature vectors of the drill cuttings example images 224, 226, and 228 (for example, the drill cuttings example images 224 and 226 may be of the first type of rock characteristics, and the drill cuttings example image 228 may be of the second type of rock characteristics) may have the same spatial relationship as the feature vector space R D The closest distance to the feature vector of the drill cuttings example image 222 in the K determined most similar drill cuttings is obtained, and the rock particles corresponding to the drill cuttings example images 224, 226, and 228 can be used for the characteristic description of the drill cuttings example image 222. When K=5, the other two feature vectors of the drill cuttings example images 230 and 232 can be included and used for the characteristic description of the drill cuttings example image 222. For example, the classification characteristics of the rock particles corresponding to the drill cuttings example image 222 can be based on the rock particles of the corresponding class included in the K determined most similar drill cuttings (K 类 ) To determine, Fig. 9 For example, in the above example, when K = 3, the first type of rock particles p 1 The ratio of (drill cuttings example images 224 and 226) is The second type of rock particles p 2 The ratio of (drill cuttings example image 228) is Warnings can be issued to inform the user when the indexed dataset is not large or diverse enough. It should be noted that although Fig.10 A two-dimensional eigenvector space is used in , but the eigenvector space R D It can be of any dimension (eg, D=1, 10, 50, ...).

[0060] In some embodiments, for each determined feature vector y k (k=1,2...k), the distance Inverse to get the corresponding score, and determine the weight The weight is the normalized score. Therefore, the smaller the distance d k Corresponding to a larger score and a larger weight, which corresponds to a higher probability of the corresponding drill cuttings. For example, the weight set of (sandstone, siltstone, shale, dolomite, limestone) is (0.9, 0.0, 0.0, 0.1, 0.0), which means that the weights (probabilities) of sandstone, siltstone, shale, dolomite, and limestone are 0.9, 0.0, 0.0, 0.1, and 0.0, respectively. The classification characteristics of the sample can be determined by selecting the most likely category using the weights (probabilities).

[0061] Return to reference Fig. 9 In process 4 of the workflow 210, by using the K most similar rock particles 216, each unknown individual drill cuttings instance image 212 (e.g., I 1 ,I 2 ,…I J ) determines the corresponding characteristic description 218 (e.g., lithology type, hardness, visual porosity, gamma ray log, porosity log). For example, the index data set in the index memory can be selected to have better query accuracy and thus better characteristic estimation accuracy. For example, in Fig. 9 In the example, index dataset A may be selected from other options such as index dataset B and index dataset C based on geographic location, geological formation, well location, depth in the well, and the like. Fig.14 2 shows the workflow of feature propagation. When K is equal to 1, the features of the retrieved drill cuttings 216 can be simply copied to the unknown individual drill cuttings instance image 212 being analyzed, such as Fig. 9 As shown. For example, when K is equal to 1, the most similar rock grain in the index may have a characteristic description as "lithology type: sandstone; hardness: 0.6, visual porosity: 5%, gamma log: 30API; porosity log: 4%...," and the unknown individual drill cuttings instance image 212 being analyzed may be determined to have the same characteristic description. When K is greater than 1, the characteristics of the retrieved drill cuttings 216 may be summarized and copied to the unknown individual drill cuttings instance image 212 analyzed using various methods. For example, for classification characteristics (such as lithology class), a probability may be calculated for each class, and the most likely class may be selected. To calculate the probability of each class, the ratio Where K 类 is the number of corresponding classes included in the number K most similar drill cuttings 216. For example, in the above Fig.10 In the example described in , when K = 3, the first type of rock particles p 1 The ratio of (drill cuttings example images 224 and 226) is The second type of rock particles p 2 The ratio of (drill cuttings example image 228) is In some embodiments, the feature vector (y j )’s weight (w j For example, in the above example, the weights of the two feature vectors corresponding to the drill cuttings example images 224 and 226 can be as shown above in Fig.10, and their sum can be used for the probability of the first class; and the weight of the feature vector corresponding to the drill cuttings instance image 228 can also be calculated and used for the probability of the second class. By using the sum of the weights of the feature vectors, the probability of each class can be determined more accurately. For example, in the case of lithology type, if the probability of (sandstone, siltstone, shale, dolomite, limestone) is (0.9, 0.0, 0.0, 0.1, 0.0), then sandstone (probability 0.9) can be assigned to the unknown individual drill cuttings instance image 212 being analyzed. In particular, with respect to lithology type, subcategories can be created based on geological hierarchies, such as Fig.11 shown.

[0062] Fig.11 is a schematic diagram showing an example of lithology types and geological layers. For example, lithology types may include a main model class 240, a main class 242 (e.g., sandstone, siltstone, mudstone, mudstone, limestone, dolomite), and a mixed probability class 244 (e.g., marl, carbonate. shale (carb. shales)). If more than one lithology type has a probability greater than a threshold (e.g., 0.3), a sub-category 236 may be created. For example, when the probabilities of (sandstone, siltstone, shale, dolomite, limestone) are (0.6, 0.4, 0.0, 0.0, 0.0), instead of simply assigning "sandstone", "silty sandstone" 246 may be assigned to the unknown individual drill cuttings instance image 212 being analyzed because both sandstone and siltstone have probabilities greater than the threshold. Similarly, when the probability of (sandstone, siltstone, shale, dolomite, limestone) is (0.4, 0.6, 0.0, 0.0), "sandy siltstone" 246 can be assigned to the unknown individual drill cuttings instance image 212 being analyzed. In addition, grain size properties can also be considered when assigning lithology names. For example, instead of assigning "sandstone", grain size can be represented by "very fine sandstone". Thus, drill cuttings described in the manual may have fixed properties from the manual description, while drill cuttings that receive propagated properties may have enhanced or adjusted property assignments. In the case of numerical properties (such as visual porosity), an average or weighted average of the properties of the K most similar drill cuttings 216 can be used (e.g., y in the target class). j The weight w j ). The analyzed drill cuttings instance images, features, and drill cuttings properties can be added to the index dataset for the next query. For example, in a geological formation, the lithology may change continuously, and drill cuttings at a certain depth of a well may be associated with drill cuttings at the same depth from the same well or a different well. Therefore, the drill cuttings properties of the analyzed drill cuttings instance images can be associated with the drill cuttings properties of other drill cuttings obtained from the same depth of the same well or a different well, such as Fig.14 shown.

[0063] Fig.12 A user interface (UI) 260 of software for external information for calcium determination and acid test is shown displayed on the electronic display 80, and a user interface 280 with multiple parameters to make the query and estimation results more accurate. For example, the user can select reference data for the best accuracy of the query. The reference data can be obtained from the analysis of adjacent wells or wells from similar formations in another location. The detectable lithology class can be used to constrain the query based on the lithology type. In many cases, the user can get clues about what lithology type can or cannot be detected on the analyzed image through external information. External information such as (1) calcium determination of the same drill cuttings sample, (2) acid test of the same drill cuttings sample, (3) image of the same drill cuttings sample under ultraviolet light, (4) local geological background, (5) well logging, etc. The process of adding constraints can be done manually by the user or automatically by the system 54. For example, the calcium determination method gives the percentage 262 of (a) calcite, (b) dolomite and (c) other minerals. For example, if the percentage of other minerals is greater than a threshold value (e.g., 90%), carbonates (calcite and dolomite) can be removed from the detectable class. On the other hand, if the percentage of other minerals is greater than a threshold value (e.g., 10%), sandstone can be removed from the detectable class. An acid test may involve dropping an acid on a single drill cutting and observing a chemical reaction. The acid test 264 can be used to distinguish between carbonates (limestone, dolomite) and debris (sandstone, siltstone) because carbonate minerals (calcite, dolomite) react with acid. This information can be used to verify or correct the lithology estimates obtained in process 116. In process 116, the acid test results can be assigned to the lithology estimates for each individual instance.

[0064] The parameter of the number of neighbors to be checked corresponds to the value K mentioned above. This value will have a certain impact on the query results. A default value will be provided, but the user can change this parameter. The parameter of the similarity threshold of the unknown class is another way to constrain the query based on the lithology type. When there are no drill cuttings similar to the analyzed drill cuttings in the index reference data set, the "unknown" category can be assigned to the analyzed drill cuttings. Similarity can be defined as the Euclidean distance or cosine distance between the feature vector of the analyzed drill cuttings and the feature vector of the drill cuttings in the reference data set. For example, when the Euclidean distance between the feature vector of the analyzed drill cuttings and the feature vector of the drill cuttings in the reference data set is less than the similarity threshold, the analysis and control system 50 can determine that there are no similar drill cuttings in the index reference data set, and the "unknown" category can be assigned to the analyzed drill cuttings. The probability threshold of the uncertain class is another way to constrain the query based on the lithology type. When K>1, the output lithology type can be expressed as the probability of each category. When there is no obvious lithology class, the "uncertain" category can be assigned to the drill cutting characteristics so that the probabilities of the possible lithology classes are all less than the first threshold (e.g., 0.4). For example, when the probability of (sandstone, siltstone, shale, dolomite, limestone) is (0.9, 0.0, 0.0, 0.1, 0.0), sandstone may be selected because it is dominant. When it is (0.3, 0.3, 0.2, 0.1, 0.1), the "uncertain" category may be output because there is no dominant class with a probability greater than a second threshold (e.g., 0.4). The threshold may be controlled by the user or automatically by the system 54 to reduce the false positive rate.

[0065] Fig.13 An output display of the estimated properties of the image 300 of the rock sample is shown. It should be noted that the calculated properties can be displayed on the user interface 260 in a variety of ways, and Fig.13 The embodiment shown is an example. For example, when a user hovers or selects (e.g., using a mouse or touch screen) drill cuttings 302, corresponding information (e.g., characteristic information (A) and the most similar drill cuttings and corresponding characteristics (B) in window 304) can be displayed. In addition, the lithology percentage of image 300 can be displayed (e.g., in (C)), and the drill cuttings image projected in 2D or 3D space after reducing the dimension from the original D-dimensional feature space using a projection technique such as t-SNE (t-distributed stochastic neighbor embedding) or PCA (principal component analysis) can also be displayed. The colors and patterns used in the percentages may be lithology patterns of the user's company standards or based on USGS (United States Geological Survey) standards. The characteristic description of drill cuttings (D) can display the distribution of lithology types (e.g., dolomite 306, limestone 308, sandstone 310, mudstone 312, uncertain category 314, unknown category 316) on image 300.

[0066] As mentioned previously, in geological formations, lithology may vary continuously, and drill cuttings at a certain depth in a well may be associated with drill cuttings from the same depth in the same well or in a different well. Therefore, the drill cuttings characteristics of the analyzed drill cuttings example image can be associated with the drill cuttings characteristics of other drill cuttings obtained from the same depth in the same well or in a different well. Fig.14 As shown, well 352 (well A) can be located at a displacement DX from well 354 (well B). Drill cuttings can be obtained from different depths of well 352 and well 354. The depth direction can be along direction 355 (e.g., Z axis), and the displacement DX can be along direction 356 (e.g., X axis). The cuttings characteristics of drill cuttings obtained from the same depth (e.g., depth 358) of wells 352 and 354 can be related. The drill cuttings obtained from depth 358 of well 352 can be analyzed and can be analyzed by using the above method in Figures 6 to 13 The method described in obtains geological descriptions and / or well logs and saves them in an indexed data set (e.g., Index A).

[0067] At step 1 of workflow 350, a cuttings image of drill cuttings obtained from depth 358 of well 354 may be compared to a cuttings image of drill cuttings obtained from depth 358 of well 352. At step 2 of workflow 350, based on the similarity between the cuttings images from different wells, cuttings properties of drill cuttings obtained from depth 358 of well 352 (e.g., stored in index A) may be used in property estimation (e.g., process 210) to obtain cuttings properties of drill cuttings obtained from depth 358 of well 354. Cuttings from multiple wells may be analyzed, and corresponding geological descriptions and / or well logs may be stored in various index data sets (e.g., index A, index B, index C) in an index memory. Based on the similarity between the cuttings images, an index data set in the index memory may be selected for query in property estimation (e.g., step 3 of process 210) to obtain better accuracy of property estimation.

[0068] In some embodiments, Fig.14 Workflow 350 in process 200 can also be used to obtain well log values ​​(e.g., in step 3 of process 200). For example, drill cuttings and well log measurements can be obtained for well 352, and drill cuttings without well log measurements can be obtained for well 354. Thus, based on the similarity of the drill cuttings from the two wells, the well log measurements of well 352 can be used for well 354.

[0069] The techniques and systems disclosed herein relate to analyzing rock drill cuttings and measuring physical lithological characteristics of rock drill cuttings. An image analysis workflow is provided, which includes multiple computing modules to automatically extract relevant geological information from rock drill cuttings. Reference data, manual instructions, and log values ​​are associated and used to determine the rock characteristics of rock drill cuttings. Software for image analysis is developed and the results are displayed in various views. The results can be used to control related equipment, such as a drilling system 10 and / or a drilling plan of the drilling system 10 based on the rock characteristics (e.g., lithology) of the rock drill cuttings. For example, the analysis and control system 50 can automatically adjust one or more operating parameters of the drilling system 10 from which the rock drill cuttings are obtained based at least in part on the rock characteristics of the determined rock drill cuttings. The techniques and methods disclosed herein can allow the acquisition of high-quality log curves for real-time and / or near real-time geological formation evaluation and geosteering, which can be used to more effectively and accurately control the drilling process. Although the examples described above are illustrated for wellbores on land, similar methods can be applied to any acquisition structure.

[0070] The specific embodiments described above have been described by way of example, and it should be understood that these embodiments may have various modifications and alternative forms. It should also be understood that the claims are not intended to be limited to the specific forms disclosed, but to cover all modifications, equivalents and alternatives that fall within the spirit and scope of the present disclosure.

[0071] In the claims, means-plus-function clauses are intended to cover structures described herein that perform the recited function, not only structural equivalents, but equivalent structures. Thus, for example, while a nail and a screw may not be structurally equivalent in that a nail employs a cylindrical surface to secure wooden parts together while a screw employs a helical surface, in the context of fastening wooden parts, a nail and a screw may be equivalent structures. It is the express intent of the applicant not to invoke 35 U.S.C. 112, paragraph 6, to impose any limitations on any of the claims herein, except for those limitations in the claims that expressly use the words "means for" and the associated function.

Claims

1. A method comprising: Receiving, via the analysis and control system, an image of drill cuttings instances of drill cuttings of a geological formation; Extracting a set of drill cutting features from the drill cutting example image via an analysis and control system, wherein the set of drill cutting features corresponds to a feature vector in a feature vector space; querying an index data set using the feature vector via an analysis and control system, wherein the index data set includes one or more reference feature vectors in the feature vector space; identifying, via the analysis and control system, a plurality of reference feature vectors from the one or more reference feature vectors, wherein each of the plurality of reference feature vectors is associated with a respective reference drill cutting; determining, via the analysis and control system, a set of drill cuttings characteristics for the drill cuttings based on reference drill cuttings associated with a plurality of reference characteristic vectors; and Via the analysis and control system, classification characteristics of the drill cuttings are determined based on the set of drill cuttings characteristics.

2. The method according to claim 1, further comprising: One or more operating parameters of a drilling system from which the drill cuttings were obtained are automatically adjusted via the analysis and control system based at least in part on the classified characteristics of the drill cuttings.

3. The method according to claim 1, wherein: selecting an index data set from the one or more index data sets based on at least one of a geographic location of the drill cuttings, a geological formation of the drill cuttings, a well location of the drill cuttings, or a depth of the drill cuttings; and The index dataset is updated by using the additional reference drill cuttings instance images.

4. The method according to claim 1, wherein the plurality of reference feature vectors: identifying using a nearest neighbor search method by comparing the corresponding distance between the feature vector and each of one or more reference feature vectors in an indexed data set in a feature vector space; and One or more classes of reference drill cuttings are corresponding, and a classification characteristic is determined based on a corresponding number of reference feature vectors in each of the one or more classes.

5. The method of claim 4, wherein a corresponding weight is calculated for each of the plurality of reference feature vectors using a corresponding distance between the feature vector and each of the plurality of reference feature vectors. 6 . The method of claim 5 , wherein a respective weighted sum is calculated for each of the one or more classes of reference drill cuttings, and wherein the classification characteristic is determined based on the respective weighted sum.

7. The method of claim 6, wherein a sub-category lithology classification is determined when the sum of at least two corresponding weights is greater than a threshold value.

8. The method of claim 4, wherein the unknown lithology classification is determined when: a respective distance between the feature vector and each of one or more reference feature vectors in the index data set is less than a threshold; and The sum of the corresponding weights is less than the threshold.

9. The method of claim 1, wherein the classification characteristic is a lithology class determined by one or a combination of grain size or physical characteristics of the drill cuttings.

10. A method comprising: extracting, via the analysis and control system, a respective set of reference drill cuttings features from each of the one or more reference drill cuttings instance images of the geological formation, wherein each of the one or more reference drill cuttings instance images is associated with a respective reference drill cutting, and wherein the respective set of reference drill cuttings features corresponds to a respective reference feature vector; determining, via the analysis and control system, a respective set of reference drill cuttings characteristics for respective reference drill cuttings based at least on the guided characterization; and Via the analysis and control system, each of the one or more reference drill cuttings instance images is indexed with a corresponding set of reference drill cuttings characteristics and a corresponding reference feature vector to generate an indexed data set.

11. The method of claim 10, wherein the respective set of reference drill cuttings properties of the respective reference drill cuttings are determined based at least on well logs associated with the respective reference drill cuttings.

12. The method of claim 10, wherein the guided characterization comprises corresponding measured values ​​of physical properties of corresponding reference drill cuttings.

13. A system comprising: a drilling apparatus configured to obtain a rock sample from a well; an image acquisition system for obtaining images of rock samples; and An analysis and control system, the analysis and control system being constructed to: Obtaining drill cuttings example images of drill cuttings from images of rock samples; extracting a set of features from the drill cuttings instance image based on the reference data; and Determine the lithology classification of drill cuttings.

14. The system of claim 13, wherein the analysis and control system is further configured to: adjusting drilling equipment based on the lithology classification of the rock samples; and A property description is generated based at least on the measured values ​​of the physical properties of the drill cuttings, wherein the property description is used to determine a lithological classification of the drill cuttings.

15. The system of claim 13, wherein the analysis and control system is further configured to display: The image depicts one or more drill cuttings instance images associated with corresponding drill cuttings characteristics; The lithology percentage of the rock sample; and A multi-dimensional map of a rock sample having a plurality of drill cuttings instance images, wherein the multi-dimensional map indicates a distribution of lithological classifications of the rock sample.

Images