WO2019167030A1 - Identification et enregistrement de propriétés d'échantillons de carotte - Google Patents

Identification et enregistrement de propriétés d'échantillons de carotte Download PDF

Info

Publication number
WO2019167030A1
WO2019167030A1 PCT/IB2019/051730 IB2019051730W WO2019167030A1 WO 2019167030 A1 WO2019167030 A1 WO 2019167030A1 IB 2019051730 W IB2019051730 W IB 2019051730W WO 2019167030 A1 WO2019167030 A1 WO 2019167030A1
Authority
WO
WIPO (PCT)
Prior art keywords
scanner
data
control unit
core samples
server
Prior art date
Application number
PCT/IB2019/051730
Other languages
English (en)
Inventor
Aaron MAHER
Vince Gerrie
Sebastian GOODFELLOW
Original Assignee
Kore Geosystems Inc.
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Kore Geosystems Inc. filed Critical Kore Geosystems Inc.
Publication of WO2019167030A1 publication Critical patent/WO2019167030A1/fr

Links

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/84—Systems specially adapted for particular applications
    • E—FIXED CONSTRUCTIONS
    • E21—EARTH DRILLING; MINING
    • E21B—EARTH DRILLING, e.g. DEEP DRILLING; OBTAINING OIL, GAS, WATER, SOLUBLE OR MELTABLE MATERIALS OR A SLURRY OF MINERALS FROM WELLS
    • E21B49/00—Testing the nature of borehole walls; Formation testing; Methods or apparatus for obtaining samples of soil or well fluids, specially adapted to earth drilling or wells
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/62—Systems in which the material investigated is excited whereby it emits light or causes a change in wavelength of the incident light
    • G01N21/71—Systems in which the material investigated is excited whereby it emits light or causes a change in wavelength of the incident light thermally excited
    • G01N21/718—Laser microanalysis, i.e. with formation of sample plasma
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N2201/00—Features of devices classified in G01N21/00
    • G01N2201/10—Scanning
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N2201/00—Features of devices classified in G01N21/00
    • G01N2201/12—Circuits of general importance; Signal processing
    • G01N2201/129—Using chemometrical methods
    • G01N2201/1296—Using chemometrical methods using neural networks

Definitions

  • the present disclosure relates generally to mineral exploration systems. More particularly, the present disclosure relates to a system for acquiring data, identifying and logging properties of core samples, identifying and logging rock properties while drilling.
  • a user may examine the core sample and manually log data, including an identification of the rock type and other rock classification properties. This may result in inconsistent or non-existent images of the core sample, as well as a subjective identification of rock classification properties of the core sample. Further, the process of logging the data is time-consuming and the data may not be readily accessible from external or other locations.
  • FIG. 1 is a schematic diagram of an example system for identifying and logging core samples, identifying and logging rock properties while drilling.
  • FIG. 2 is a schematic diagram of a scanner of the example system of Fig. 1
  • FIG. 3 is a schematic diagram of a control unit of the example system of Fig. 1 .
  • FIG. 4 is a schematic diagram of further example system for identifying and logging core samples.
  • FIG. 5 is a block diagram of a system for standardizing outputs between two devices.
  • Fig. 6 is a schematic diagram of a scanning control interface in the system of Fig. 1 .
  • Fig. 7 is a schematic diagram of a depth referencing interface in the system of Fig. 1 .
  • Fig. 8 is a schematic diagram of a lithology interface in the system of Fig. 1 .
  • Fig. 9 is a flowchart of a method of performing image analysis in the system of Fig. 1 .
  • Fig. 10 is a schematic diagram of the image analysis during the method of Fig. 9.
  • the present disclosure provides a system for identifying and logging properties of core samples and identifying and logging rock properties while drilling.
  • the system includes a server, a scanner and a control unit.
  • the control unit is configured to connect to a drill rig to extract rock property data through drilling parameters.
  • the scanner is configured to scan the core samples using optical cameras or other scanning
  • the server, the scanner and the control unit are interconnected, for example via a network, or via a portable device configured to connect to each of the server, the scanner and the control unit individually, to exchange and synchronize data.
  • the server, the scanner, and the control unit each include a respective processor configured to employ an image analysis subsystem capable of extracting optical features such as color, shape, and texture, and applying machine learning models to generate predicted rock classifications, such as lithology, alteration, and geotechnical parameters.
  • Application of the image analysis subsystem will allow for delivery of reports, visualizations, and automatic classification of core images.
  • the system provides consistent, high quality imaging, logging, rock classification and element analysis, as well as an intelligent synchronization system to ensure that data is not lost or duplicated, and that all data is eventually synchronized to a database on the server.
  • Fig. 1 depicts a system 100 for identifying and logging core samples 101 extracted from a drill hole 102 by a drill rig 105.
  • the drill rig 105 includes a base unit configured to anchor the drill rig on the ground and to support other components of the drill rig 105.
  • the base unit can include trusses, beams, cables, supporting surfaces and the like.
  • the drill rig 105 includes a drill bit configured to drill the drill hole 102.
  • the drill bit may be connected to one or more operative components such as motors, pulleys, winches, and the like supported on the base unit for controlling operation of the drill bit (e.g. to raise, lower and actuate the drill bit).
  • the drill bit may be connected to the one or more operative components via an overshot line.
  • the drill rig 105 further includes a sampler coupled to the overshot line and the drill bit and configured to extract the core samples 101 from the drill hole 102.
  • the drill rig 105 further includes a rig controller operatively coupled to the components of the drill rig 105, including the drill bit and the sampler to control the functions and parameters of the drill rig 105.
  • the rig controller may control drilling operations, sampling operations, and the like.
  • the rig controller may further control parameters such as the speed or direction of the drill bit, or the like.
  • the rig controller may further be configured to obtain, store and process data from the drill bit, from the sampler, and from other components of the drill rig.
  • the system 100 further includes a server 1 10, a scanner 120, and a control unit 130.
  • the control unit 130 is coupled to the drill rig 105 to provide external control of operation of the drill rig 105.
  • the scanner 120 is configured to scan the core samples 101 for identifying and logging properties of the core samples 101 .
  • the server 1 10 is configured to store and process data from the scanner 120 and the control unit 130.
  • the server 1 10, the scanner 120 and the control unit 130 are mutually coupled by a network 140 for data communications.
  • suitable networks include internet protocol (IP) networks, such as intranet, a local-area network, a wide-area network, a virtual private network (VPN), a Wi-FiTM network, a short-range wireless network (e.g.
  • BluetoothTM or BluetoothTM Low Energy the internet, combinations of such, and similar.
  • Fig. 2 depicts a schematic diagram of the scanner 120.
  • the scanner 120 includes a main body 200 configured to receive a core sample 101 for scanning, one or more input devices and one or more output devices, generally indicated as an input/output device 204.
  • the scanner 120 further includes one or more cameras 210 disposed in the main body configured to scan images of the core sample 101 , a scanner processor 220 configured to control the operation of the scanner 120, and a scanner communication module 230 configured for data communications, for example, via network 140.
  • the scanner processor 220 and the scanner communication module 230 may be internal to the scanner 120.
  • the main body 200 is configured to receive a core sample 101 for scanning.
  • the main body 200 defines a chamber 202 in which the core sample is received.
  • the main body 200 may further include a lid enclosing the chamber 202.
  • the chamber 202 may be configured to receive a core box containing multiple core samples 101 .
  • the scanner 120 may include a drawer in which a core box is placed to be received in the chamber 202 for scanning.
  • the input and output devices 204 are configured to receive commands for controlling the operation of the scanner, receive input from a user to add or modify data, and to display results, data parameters, or other information.
  • the input and output devices 204 are interconnected with the scanner processor 220 to execute the commands.
  • the input and output devices 204 therefore include any suitable
  • the input and output devices 204 may form a scanner user interface configured to display images of core samples, display outputs from algorithms, provide visual and audible event notifications and allow a user to control scanning and data entry operations.
  • the scanner user interface may provide a scanning control interface 600, as depicted in FIG. 6, configured to allow a user to control scanning operations.
  • the scanning control interface may provide a scan initiation interface 602 including a start button, to initiate a scanning operation.
  • the scanning operation may be initiated by indicating a project identifier and a box identifier (e.g. via manual input by the user).
  • a new scanning operation may be initiated.
  • the scan initiation interface 602 may further provide options to select parameters for conducting the scan (e.g. type of technology used for scan, wet/dry conditions for scan, and the like).
  • the scan initiation interface 602 may accept input from the user to name or rename the scanned image(s).
  • the scanning control interface 600 may further provide a cropping tool 610 to allow the user to crop the scanned image to the desired or usable area for further processing or analysis.
  • the cropping tool 610 may allow adjustment of the top, bottom, left and right edges of the image, for example, by dragging and dropping respective crop lines 612 (depicted in dashed lines) within the image.
  • the cropping tool 610 may include sliders, text boxes for receiving crop values, or other suitable interfaces for cropping the image.
  • the scanning control interface 600 may display the crop lines 612 superimposed over the scanned image to allow the user to visualize the cropped image.
  • the scanning control interface 600 may further provide a template definition tool 620 to allow the user to define a template 622 based on the arrangement of core samples 101 within a core box.
  • the template 622 (depicted in dash-dot lines) may be defined based on a number of rows, an x-value (i.e. defining the horizontal starting point of the rows), a y-value (i.e. defining a vertical starting point of the rows), a length of the rows, a height of the rows, and a spacing between rows.
  • the template definition tool 620 may include sliders 624 for adjusting the above-mentioned parameters.
  • the template definition tool 620 may include text boxes, drag and drop mechanisms, or other suitable interfaces for adjusting the above- mentioned parameters. Further, the scanning control interface 600 may display the template 622 superimposed over the scanned image to allow the user to easily visualize the template with respect to the underlying core box. The template definition tool 620 may further allow the user to save the template 622 as a preset for application to future scanned images. [0022] The scanning control interface 600 may further provide a depth reference tool 630 to allow the user to define a starting depth of the core box, as well as a total depth for the core box. For example, the total depth may be a maximum length of core containable in the core box. Accordingly, as is described further below, subsequent interfaces or image analysis techniques may determine an actual length of core contained in the core box based on the maximum length, and a ratio of core to non-core lengths within the box.
  • the scan initiation interface 602, the cropping tool 610, and the template definition tool 620 of scanning control interface 600 are depicted as being displayed simultaneously and controllable within a single view.
  • the scanning control interface 600 may present the user with each tool sequentially, for example, using different tabs, pop-up windows, or the like, to guide the user through the scanning operation.
  • the scanner user interface may further provide a depth referencing interface 700 configured to allow a user to input and alter depth references and measurement for further processing and use in image analysis.
  • the depth referencing interface 700 may allow a user to select non-rock portions, for example, by clicking and dragging boxes 702 to identify the non-rock portions. In some examples, the user may simply drag across a horizontal length.
  • the row may be identified based on the location of the initial click and the template of the image (e.g. as identified using the template definition tool 620). Further, the upper and lower bounds of the row may be determined based on the template of the image.
  • the depth referencing interface 700 may further allow a user to add a reference depth 704 to a box 702, thereby identifying the contents of the box as a depth marker and indicating the depth.
  • a first reference depth 704 the depths of other boxes 702 may also be computed based on the proportions and locations of the boxes 702 and the known total depth of the core box (e.g. as identified using the depth reference tool 630).
  • the depth referencing interface 700 may further provide options to delete, move, adjust or otherwise alter the boxes 702.
  • the scanner user interface may further provide a lithology interface 800 configured to allow a user to input and alter lithology
  • the lithology interface 800 may allow a user to select rock portions, for example, by clicking and dragging segments 802 to identify the rock portions. In some examples, the user may simply drag across a horizontal length, and the row, upper and lower bounds of the row may be determined based on the template of the image. In some examples, the segments 802 may span multiple rows.
  • the lithology interface 800 may further allow a user to select a lithology classification code 804 for each segment. For example, the classification code 804 may be selected from a list of options (e.g. buttons, dropdown lists, or the like), or may be manually input.
  • the lithology interface 800 may further provide options to split or merge segments 802.
  • the camera 210 is disposed in the main body 200 and is configured to scan images of the core samples 101 within the chamber 202.
  • the camera 210 is an optical camera (i.e. capturing visible light) configured to capture optical images.
  • the scanner 120 may include a series of optical cameras 210.
  • the scanner 120 may further include a laser- induced breakdown spectroscopy (LIBS) module 214 disposed in the main body 200 and configured to scan the core samples 101 within the chamber 202 for LIBS analysis.
  • LIBS laser- induced breakdown spectroscopy
  • other suitable technology capable of scanning or capturing images of the core samples for analysis may be provided.
  • the scanner processor 220 may include a central-processing unit (CPU), a microcontroller, a microprocessor, a processing core, a field programmable gate array (FPGA), or similar.
  • the scanner processor 220 may include multiple cooperating processors.
  • the scanner processor 220 may cooperate with a non-transitory computer readable medium such as a scanner memory (not shown) to execute instructions to realize the functionality discussed herein.
  • the scanner processor 220 is interconnected with the optical camera 210 to receive and process image data. For example, where the scanner 120 includes a series of optical cameras 210, the scanner processor may combine the images from the series of optical images 210 to form a single optical image via one or more stitching
  • the image data including individual image data, and the combined stitched image data, may be stored in the scanner memory for further processing.
  • the scanner processor 220 is further configured to apply image analysis to the images or the image data from the optical camera 210 to identify properties of the core samples 101 .
  • the scanner processor 220 may employ an image analysis subsystem capable of extracting optical features, such as color, shape, and texture, and applying machine learning models to generate predicted rock classifications such as lithology, alteration, and geotechnical parameters.
  • the scanner processor 220 may be configured to calculate depth measurements from the images.
  • the optical image features and classifications may also be stored in the scanner memory for further processing.
  • the image analysis subsystem may be configured to perform a method 900 for classifying a core sample.
  • the method 900 may be applied to core samples in a core box having one or more rows of core samples.
  • the image analysis subsystem selects a row of the core box for analysis.
  • the row may be detected, for example, based on user input (e.g. via a template selection at the user interface), based on a pre-applied template, based on a machine learning analysis of the image data, or other suitable methods for identifying rows within a core box.
  • the image analysis subsystem may process the image 1000 to select a row 1004 from the core box.
  • the image analysis subsystem applies a segmentation model to the row selected at block 905.
  • the segmentation model is configured to identify rock, non-rock, and depth marker segments of a row in a core box.
  • the segmentation model is a 20-layer deep convolutional neural network. It may be built using the Google TensorflowTM API.
  • the segmentation model may include 2D convolutional filters, fully connected layers, batch normalization, dropout, max-pooling, and other available elements.
  • the segmentation model may employ lnception /4 architecture, which was designed for square aspect ratio images, and adapted for a rectangular aspect ratio of the images of the core samples.
  • the segmentation model may segment the core row 1004 into pixels 1010 and may output, for each pixel 1010, a binary classification (e.g. rock/non-rock) according to You Only Look Once (YOLO) object detection.
  • a binary classification e.g. rock/non-rock
  • the segmentation may further classify each non rock value as a second binary value (e.g. depth marker/non-depth marker).
  • the segmentation model may provide each row in the core box with a mapping of each pixel 1010 to a rock/depth-marker/non-rock classification value.
  • Other suitable neural network configurations or machine learning algorithms are also contemplated.
  • the image analysis subsystem crops the row, and in particular, the segments of pixels identified as containing rock, into cropped segments.
  • adjacent cropped segments may overlap.
  • the pixels 1010 may be assigned cropped segments 1020.
  • the image analysis subsystem applies a classification model to each of the cropped segments.
  • the classification model is configured to classify rock properties, such as lithography, alteration, or the like, of the core sample.
  • the classification model is a 16-layer deep convolutional neural network built using the Google TensorflowTM API.
  • the classification model may include 2D convolutional filters, fully connected layers, batch normalization, dropout, max-pooling, and other elements.
  • the classification model may employ VGG16 architecture, which was designed for square aspect ratio images, and adapted for a rectangular aspect ratio of the images of the core samples.
  • the classification model may classify rock properties of each of the cropped segments obtained at block 915.
  • the classification model may receive, as input, the image data representing respective portions of the core sample corresponding to the cropped segments 1020, and may output a classification label (e.g. rock type, alteration, etc.).
  • the classification model may further output a confidence interval associated with the classification label.
  • the classification model may output more than one classification label and associated confidence interval.
  • the segmentation model and the classification model may be trained on different datasets.
  • a trained model exists as a TensorflowTM model file containing the trained weights and parameters (e.g. the weights of the convolution filters).
  • the models learn the parameter weights as a part of an optimization process.
  • hyperparameters such as dropout rate, batch size, and the like, may be selected. Hyperparameters may be chosen on a per model basis using a random or grid search.
  • the image analysis subsystem combines the classification results of each of the cropped segments to obtain an overall property of the core sample in the row.
  • the image analysis subsystem may provide one or more classification labels and associated confidence intervals for the entire row of core sample.
  • the image analysis subsystem may further identify a suggested transition line. For example, referring to FIG.
  • the image analysis subsystem may identify a suggested transition line 1030 located within the intervening cropped segment 1020.
  • the suggested transition line 1030 may vary based on the relative confidence intervals of class A and class B in the intervening cropped segment 1020.
  • the scanner processor 220 may also be interconnected with the LIBS module 214 to receive and process LIBS data.
  • the scanner processor 220 is further configured to apply image analysis based on the LIBS data from the LIBS module 214 to identify further properties of the core samples 101 .
  • the scanner processor 220 may employ a further image analysis subsystem capable of extracting optical features, such as color, shape, and texture, and applying further machine learning models to generate predicted rock classifications such as lithology, and alteration.
  • the optical image features and classifications may also be stored in the scanner memory for further processing.
  • the scanner processor 220 may apply a preliminary image analysis to the image data from the optical camera 210 to identify certain portions of the core sample 101 for further analysis.
  • the scanner processor 220 may control the LIBS module 214 to scan the identified portions of the core sample 101 for LIBS analysis.
  • the scanner processor 220 is further configured to generate reports.
  • the reports may include data such as the rock classification properties of the core sample.
  • the reports may also identify errors or problems, such as quality of the images.
  • the reports may be displayed to a user via the input/output device 204. Further, the user may review and address the data and the errors or problems by interacting with the input/output device 204. For example, the user may correct rock classification properties, or add new rock classification properties.
  • the scanner processor 220 updates the reports and may store the reports in the scanner memory for further processing.
  • the scanner processor 220 is also interconnected with the scanner
  • the scanner communication module 230 is configured for bidirectional data communications, for example through network 140, and may include suitable hardware (e.g. transmitters, receivers, network interface controllers and the like) allowing the scanner 120 to communicate with other computing devices, such as the server 1 10 and the control unit 130.
  • the specific components of the scanner communication module 230 are selected based on the type of network or other links that the scanner 120 is required to communicate over.
  • the scanner 120 may send and receive the image data, the rock classification properties, and the reports to the server 1 10 and the control unit 130 via the network 140 using the scanner communication module 230. For example, rather than performing image analysis at the scanner processor 220, the scanner 120 may send raw image data via the scanner
  • the scanner 120 and the control unit 130 may communicate via the network 140, while in other embodiments, the scanner 120 and the control unit 130 may communicate via a separate drill site network, such as a network specifically configured for downhole communications.
  • the scanner 120 may further include a rotation mechanism 240 configured to rotate the core sample 101 within the main body 200.
  • the rotation mechanism 240 may include a pair of rollers on which the core sample 101 rests.
  • Other suitable rotation mechanism 240 may include a pair of rollers on which the core sample 101 rests.
  • the scanner processor 220 may control the rotation mechanism 240 according to pre-defined settings. In other embodiments, scanner processor 220 may control the rotation mechanism 240 to rotate the core sample 101 according to user input received at the input device 204.
  • Fig. 3 depicts a schematic diagram of the control unit 130.
  • the control unit 130 includes a drill rig interface 300 configured to connect the control unit to the controller of the drill rig 15, a control unit processor 310 to control the operations of the control unit 130, a control unit memory 320, and a control unit communication module 330 configured for bidirectional data communications.
  • the drill rig interface 300 is configured to connect the control unit 130 to the controller of the drill rig 15.
  • the interface 300 may include suitable hardware (e.g.
  • the communication link between the interface 300 and the controller may be either wired or wireless, or a combination of wired and wireless.
  • the interface 300 and the controller may each include a port for universal serial bus (USBTM) communications or the like.
  • USBTM universal serial bus
  • the control unit processor 310 may include a central-processing unit (CPU), a microcontroller, a microprocessor, a processing core, a field programmable gate array (FPGA), or similar.
  • the control unit processor 310 may include multiple cooperating processors.
  • the control unit processor 310 may cooperate with a non-transitory computer readable medium such as the control unit memory 320 to execute instructions to realize the functionality discussed herein.
  • the control unit processor 310 is interconnected with the interface 300 to give commands and comments to the drill rig controller of the drill rig 15.
  • the control unit processor 310 may be configured to control the functions and parameters of the drill rig 15 via the interface 300.
  • the control unit processor 310 may also monitor and record functions and parameters of the drill rig 15 via the interface 300.
  • the control unit processor 310 may control the drill bit to drill the hole further, or may monitor the drilling action of the drill rig 15.
  • the control unit processor 310 may control the speed and direction of the drill bit, and/or may monitor and record the speed and direction of the drill bit.
  • the control unit processor 310 is further configured to analyze the functions and parameters of the drill rig to determine rock properties.
  • control unit processor 310 may analyze the speed of the drill bit and the depth of the drill hole to extrapolate rock properties such as hardness.
  • control unit processor 310 may monitor and record the functions and parameters of the drill rig 15 in real-time.
  • control unit processor 310 may be configured to download functions and parameters of the drill rig from an earlier drilling period. The functions and parameters of the drill rig as recorded or downloaded by the control unit processor 310, as well as the rock properties may be stored in the control unit memory 320 as drill rig data for further processing.
  • control unit processor 310 may be further configured to apply image analysis to images from the camera 210 to identify properties of the core samples 101 .
  • the control unit 130 may receive optical images from the scanner 120 via the network 140 and the control unit processor 310 may employ a machine learning subsystem to the optical images to predict the rock classification properties such as lithology, structure and the like.
  • the control unit 130 may receive LIBS data from the LIBS module 214 and the control unit processor 310 may employ a further machine learning subsystem to identify further rock classification properties, such as elemental analysis and the like.
  • the control unit 130 may then send the results of image analysis to the scanner 120 via the network 140.
  • the rock classification properties may be stored in the control unit memory 320 for further processing.
  • the control unit processor 310 is further configured to generate reports.
  • the reports may include data such as the rock classification properties of the core sample and rock classification properties based on drill rig data.
  • the reports may also identify errors or problems, such as quality of the images.
  • the control unit processor 310 may store the reports in the control unit memory 320 for further processing.
  • the control unit processor 310 is interconnected with the control unit communication module 330.
  • the control unit communication module 330 is configured for bidirectional data communications, or example through the network 140, and may include suitable hardware (e.g. transmitters, receivers, network interface controllers and the like) allowing the control unit 130 to communicate with other computing devices, such as the server 1 10 and the scanner 120.
  • control unit communication module 330 The specific components of the control unit communication module 330 are selected based on the type of network or other links that the control unit 130 is required to communicate over.
  • the control unit may send and receive data such as image data, and the rock classification properties to the server 1 10 and the scanner 120 via the network 140 using the control unit communication module 330.
  • the server 1 10 includes a server processor 1 12 configured to control the operations of the server 1 10, and a database 1 14.
  • the server processor 1 12 may include a central-processing unit (CPU), a microcontroller, a microprocessor, a processing core, a field programmable gate array (FPGA), or similar.
  • the processor may include multiple cooperating processors.
  • the server processor 1 12 may cooperate with a non-transitory computer readable medium such as a server memory (not shown) to execute instructions to realize the functionality discussed herein.
  • the server memory may include a combination of volatile (e.g.
  • RAM Random Access Memory
  • ROM read only memory
  • EEPROM Electrically Erasable Programmable Read Only Memory
  • flash memory All or some of the server memory may be integrated with the server processor 1 12.
  • the server memory stores computer readable instructions for execution by the server processor 1 12.
  • the server memory also stores a database 1 14 for logging the properties of core samples and parameters and functions of the drill rig.
  • the database 1 14 may store raw data, such as the image data, drill rig functions and parameters and the like, as well as processed data, such as the rock classification properties, the reports and the like.
  • the server 1 10 may be configured to receive the raw data and process the raw data prior to storing both the raw data and the processed data, while in other embodiments, the server 1 10 may receive both the raw data and the processed data for storing in the database 1 14.
  • the system 100 with one or more servers 1 10 is implemented as a cloud-based service.
  • Commercially available data storage and processing resources which may be known as cloud services, may be configured with the functionality described herein.
  • the server 1 10 may also include input devices (not shown) interconnected with the server processor 1 12, such as a keyboard and mouse, as well as output devices (not shown) interconnected with the server processor 1 12, such as a display.
  • the input and output devices can be connected to the server processor 1 12 via a server network interface and another computing device. That is, the input and output devices can be local to the server 1 10 or remote from the server 1 10.
  • the server 1 10 may thus display images of core samples and reports to allow a user to visualize core samples and parameters.
  • the server processor 1 12 is configured to manage components of the system 100. In particular, the server processor 1 12 may manage instrumentation and
  • the server processor 1 12 may manage licenses of the scanner 120, the control unit 130, the sampling module, or other components of the system 100 to enable or disable functionality.
  • the server processor 1 12 may further manage software updates for the components of the system 100, as well as the download of algorithms onto the scanner 120, the control unit 130, or other components of the system 100.
  • the server processor 1 12 may further be configured to manage error and maintenance notifications.
  • the server processor 1 12 may interface with the server memory to manage inventory.
  • the server memory may store an association of an asset to asset parameters, such as an asset location, an asset renter, an asset owner, and the like.
  • the server processor 1 12 may further manage data across the system 100, for example, to ensure no data is lost, and to prevent duplication of data.
  • the server processor 1 12 is further configured to apply image analysis to the images from the camera 210 to identify properties of the core samples.
  • the server 1 10 may receive optical images from the scanner 120 via the network 140, and the server processor 1 12 may employ the image analysis subsystem capable of extracting optical features, such as color, shape, and texture, and applying machine learning models to generate predicted rock classifications such as lithology, alteration, and geotechnical parameters.
  • the server processor 1 12 may apply, sequentially, the segmentation model and the classification model to the received optical images.
  • the server 1 10 may store, in the server memory, multiple separate segmentation models and classification models according to site parameters (i.e.
  • the server 1 10 may employ a single segmentation model applicable to images of core samples regardless of site parameters, and a unique classification model based on site parameters.
  • each project, location, client, combination of such, or the like may have a unique classification model according to the desired classification groupings, names, and classification types.
  • one classification model may be configured to classify rock types in more broad categories, while another may be more granular.
  • one classification model may be configured to classify alterations according to a certain weightings of optical and LIBS image data, while other classification models may use only optical data, different weightings, or may classify rock type only.
  • the server 1 10 may receive LIBS data from the LIBS module 214 and the server processor 1 12 may employ the geochemistry analysis subsystem capable of processing geochemical LIBS data and applying machine learning models to generate predicted rock classifications such as lithology and alteration.
  • the server 1 10 may send the results of the image analysis to the scanner 120.
  • the rock classification properties may be stored in the database 1 14 for further processing.
  • the scanner 120 and the server 1 10 may cooperate to efficiently produce highly accurate results.
  • the scanner 120 provides consistent, high quality images of the core samples 101 .
  • the scanner 120 may send the results to the server 1 10 for image analysis.
  • the server 1 10, may apply the appropriate segmentation and classification models to identify the rows of core, the rock and non rock pixels, and the classification of each cropped segment and/or each row of core.
  • the results may be displayed at the scanner 120.
  • the depth referencing and lithology interfaces of the scanner 120 may display the results for verification or additional input by a user. In other examples, the scanner 120 may simply display the results without requesting verification.
  • the verified results may be stored at the server 1 10 and used to validate or improve the segmentation and classification models. Further, any data input by the user may be automatically stored at the server 1 10, including computations of relative depth measurements, starting points, ending points, and the like, based on the detected rock and non-rock pixels and the template of the image, without need for manual input of such data by the user.
  • the server processor 1 12 is further configured to analyze data and generate reports.
  • the server processor 1 12 may analyze various types of data, including optical images and LIBS data, and apply machine learning algorithms and models.
  • the reports may include data such as the rock classification properties of the core sample.
  • the reports may also identify errors or problems, such as quality of the images.
  • the server processor 220 may store the reports in the database 1 14 for further processing.
  • the scanner 120 provides consistent, high quality images of the core samples 101 and is highly efficient. For example, at a mining site in Chile, a system scanned approximately 25,000 meters of core samples in two months. The system provided predictable productivity and automatic logging of images, data, and rock classification properties. [0060] Further, by applying image analysis at any one of the scanner 120, the server 1 10 and the control unit 130, the system 100 provides consistent identification of rock classification properties, including elemental analysis. For example, a classification model was trained using roughly 6000 meters of core data from multiple drill holes from a property in Chile drilled before 2019.
  • the training dataset was composed of cropped segments identified as one of nine unique rock types: volcanic rock, quartz diorite F, diorite X, quartz diorite G, diorite, intrusion breccia, andesite-basalt, sedimentary rock, and igneous breccia.
  • Rock types identified in fewer than one thousand cropped segments were discarded.
  • the model was trained on 70% of the dataset, validated on 15% of the dataset, and tested on the remaining 15% of the dataset. Table 1 depicts the cross between the predicted label as identified by classification model against the true label.
  • the classification model had an 88% accuracy across individual cropped segments. Further, when considering the top two identified rock types based on the highest confidence intervals, the classification model had a 96% accuracy across individual cropped segments. The mean F1 score of the classification model across individual cropped segments was 84%.
  • the classification model had a 91 % accuracy when identifying rock types in a row (i.e. after combining results of several cropped segments). Further, when considering the top two identified rock types based on the highest confidence intervals, the classification model had a 98% accuracy when identifying rock types in a row. The mean F1 score of the classification model when identifying rock types in a row was 88%.
  • Fig. 4 depicts a system 400 including a server 410, a scanner 420, and a control unit 430, which are similar to the server 1 10, the scanner 120, and the control unit 130 respectively.
  • the system 400 further includes an orientation module 440, a data shuttle module 450, and a portable device 460.
  • the orientation module 440 is configured to be secured to the core sample 101 for orienting the core sample 101 when it is removed from the drill hole 102.
  • the orientation module 440 is configured to interface with other components of the system 400, such as the data shuttle module 450 and the scanner 420 to transmit orientation data.
  • the orientation module 440 interfaces with the scanner 420, for example via the scanner communication module to transmit orientation data.
  • the scanner processor may process the orientation data and control the rotation mechanism to rotate the core samples 101 according to the orientation data.
  • the orientation module 440 may include a visual indicator for a user to manually rotate the core samples 101 to be correctly oriented within the scanner 420.
  • the orientation data may be stored in the scanner memory and may be transmitted to other components of the system 400, such as the database in the server 410 for further processing.
  • the orientation data may be stored in association with the optical images or image data from the scanner 420 to provide further context for the images.
  • the orientation module 440 may include further sensors, such as a natural gamma sensor.
  • the further sensors may record data in addition to the orientation data collected by the orientation module.
  • the orientation module 440 including the further sensors may be secured to the drill bit of the drill rig to allow continuous data collection at the drill bit.
  • the data shuttle module 450 is configured to be movable along the overshot line of the drill rig 15 for communicating data between one or more downhole sensors and the control unit 430.
  • the data shuttle module 450 may be controlled by the control unit 430 to move it along the overshot line.
  • the control unit 430 may control the speed of incline or decline based on drill rig data.
  • the data shuttle module 450 may interface with the downhole sensors to receive sensor data when the data shuttle module 450 is nearby.
  • the data shuttle module 450 may connect wirelessly, optically or acoustically to receive sensor data. When the data shuttle module 450 is near the control unit 430, the data shuttle module 450 may similarly connect wirelessly, optically or acoustically to transmit the sensor data to the control unit 430.
  • the data shuttle module 450 may further include a sensor 452 configured to take measurements to map out the drill hole.
  • the sensor 452 may be integrated with the data shuttle module 450, or it may be coupled to the data shuttle module 450 via a modular interface.
  • the sensor 452 may be a gyroscope configured to take measurements as the data shuttle module 450 moves along the overshot line, thereby providing data for mapping out the drill hole.
  • the portable device 460 may be a smartphone, tablet computer, or the like, and is configured to synchronize data from the scanner 420, the control unit 430 and the server 410.
  • the portable device 460 may synchronize data between the components of the system 400 where the components are connected to different networks, or are not able to connect to the network, and thus unable to communicate directly.
  • the portable device 460 may connect to the scanner 420 via a drill-site network or via a wired communication link to receive data. The portable device 460 may then leave the drill-site network or be disconnected from the wired
  • the portable device may then transmit the data via the internet network to the server 410.
  • the data may include raw data, such as image data, orientation data and the like, or processed data, such as the rock classification properties, reports and the like.
  • the data may further include: sensor data, images, analytics, metadata, firmware upgrades, algorithm updates, error logs, and data synchronization information.
  • the system 400 may manage identical data stored on each component of the system 400 to ensure that no data is lost or duplicated, and that all data is eventually stored in the database on the server 410.
  • Fig. 5 depicts a system 500 including an interface 510 configured to standardize data communication between a first device 520 and a second device 530.
  • the first device 520 and the second device 530 may include one of: a core scanner, a drill rig control unit, an orientation module, and a data shuttle.
  • the first device 520 produces a first output 522 having a first identifier
  • the second device 530 produces a second output 532 having a second identifier.
  • the interface 510 is configured to receive and process the first output 522 and the second output 532 to produce outputs in a standardized format.
  • the interface 510 may produce an output which identifies and differentiates the first output 522 and the second output 532 via the first identifier and the second identifier.
  • the first identifier may be a device identifier for the first device 520
  • the second identifier may be a device identifier for the second device 530.
  • the system 500 allows the first device 520 and the second device 530 to communicate data via the interface 510.
  • the interface 510 may allow communications from the first device 520 and the second device 530 to other components of the system 500.
  • a system for identifying and logging core samples includes a control unit, a scanner and a server.
  • the control unit is configured to connect to a drill rig to extract core samples from a drill hole.
  • the scanner provides consistent, high quality images of the core samples, as well as automatically logging the images and rock classification properties. Image analysis to identify the rock classification properties may be applied at any one of the scanner, the server and the control unit.
  • the system 100 provides consistent, objective identification of rock classification properties, and does not require users to log the data.
  • the components of the system are interconnected to exchange and synchronize data; all data is eventually synchronized to a database on the server.

Abstract

La présente invention concerne un système donné à titre d'exemple pour identifier et enregistrer les propriétés d'échantillons de carotte extraits d'un trou de forage par un appareil de forage qui comprend : un serveur pour enregistrer les propriétés d'échantillons de carotte et gérer les instruments et les équipements ; un dispositif de balayage pour balayer les échantillons de carotte pour générer des images des échantillons de carotte ; et une unité de commande configurée pour : raccorder l'unité de commande à un dispositif de commande de l'appareil de forage ; surveiller et enregistrer des fonctions et des paramètres de l'appareil de forage ; et donner des commandes et des commentaires à l'appareil de forage. Le serveur, le scanner et l'unité de commande sont configurés pour communiquer sur un réseau pour transférer et enregistrer des données. Le dispositif de balayage et le serveur sont configurés pour appliquer une analyse d'image aux images des échantillons de carotte pour identifier des propriétés des échantillons de carotte et générer des rapports.
PCT/IB2019/051730 2018-03-02 2019-03-04 Identification et enregistrement de propriétés d'échantillons de carotte WO2019167030A1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
US201862637802P 2018-03-02 2018-03-02
US62/637,802 2018-03-02

Publications (1)

Publication Number Publication Date
WO2019167030A1 true WO2019167030A1 (fr) 2019-09-06

Family

ID=67804882

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/IB2019/051730 WO2019167030A1 (fr) 2018-03-02 2019-03-04 Identification et enregistrement de propriétés d'échantillons de carotte

Country Status (1)

Country Link
WO (1) WO2019167030A1 (fr)

Cited By (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110715837A (zh) * 2019-10-25 2020-01-21 中国石油化工股份有限公司 岩心制造方法
CN111007064A (zh) * 2019-12-13 2020-04-14 常州大学 一种基于图像识别的录井岩性智能识别方法
CN111144225A (zh) * 2019-12-03 2020-05-12 中铁第四勘察设计院集团有限公司 一种地质勘探岩芯的识别方法、装置和存储介质
CN113138178A (zh) * 2021-04-15 2021-07-20 上海海关工业品与原材料检测技术中心 一种识别进口铁矿石品牌的方法
CN113216946A (zh) * 2021-05-11 2021-08-06 西南石油大学 一种基于岩屑和元素录井数据的井壁崩落、掉块检测方法
WO2022031839A1 (fr) * 2020-08-07 2022-02-10 Bly Ip Inc. Systèmes et procédés d'analyse améliorée de carotte
CN115410049A (zh) * 2022-10-31 2022-11-29 中国石油大学(华东) 一种岩体溶蚀程度的分类评估方法及装置
WO2023132935A1 (fr) * 2022-01-07 2023-07-13 Schlumberger Technology Corporation Systèmes et procédés permettant de segmenter des instances de particules rocheuses
WO2023209270A1 (fr) * 2022-04-30 2023-11-02 Lumo Analytics Oy Procédé et système d'analyse de la teneur en éléments de carottes de forage

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20050194134A1 (en) * 2004-03-04 2005-09-08 Mcgregor Malcolm D. Downhole formation sampling
CA2793811A1 (fr) * 2007-01-29 2008-08-07 Dmitriy Repin Systeme et procede pour realiser des operations de forage de champ petrolifere en utilisant des techniques de visualisation
US20080208475A1 (en) * 2007-02-25 2008-08-28 George Karr Drilling collaboration infrastructure
US20100324868A1 (en) * 2009-06-22 2010-12-23 Russell Mark C Core Sample Preparation, Analysis, And Virtual Presentation
WO2015189286A1 (fr) * 2014-06-10 2015-12-17 Centre National D'etudes Spatiales Dispositif d'analyse spectroscopique de carottes de forage
US20160187313A1 (en) * 2014-12-31 2016-06-30 Ge Energy Oilfield Technology, Inc. Core Sample Analysis System with Conditioned Environment
US20180119535A1 (en) * 2015-05-08 2018-05-03 Schlumberger Technology Corporation Real time drilling monitoring

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20050194134A1 (en) * 2004-03-04 2005-09-08 Mcgregor Malcolm D. Downhole formation sampling
CA2793811A1 (fr) * 2007-01-29 2008-08-07 Dmitriy Repin Systeme et procede pour realiser des operations de forage de champ petrolifere en utilisant des techniques de visualisation
US20080208475A1 (en) * 2007-02-25 2008-08-28 George Karr Drilling collaboration infrastructure
US20100324868A1 (en) * 2009-06-22 2010-12-23 Russell Mark C Core Sample Preparation, Analysis, And Virtual Presentation
WO2015189286A1 (fr) * 2014-06-10 2015-12-17 Centre National D'etudes Spatiales Dispositif d'analyse spectroscopique de carottes de forage
US20160187313A1 (en) * 2014-12-31 2016-06-30 Ge Energy Oilfield Technology, Inc. Core Sample Analysis System with Conditioned Environment
US20180119535A1 (en) * 2015-05-08 2018-05-03 Schlumberger Technology Corporation Real time drilling monitoring

Cited By (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110715837A (zh) * 2019-10-25 2020-01-21 中国石油化工股份有限公司 岩心制造方法
CN111144225A (zh) * 2019-12-03 2020-05-12 中铁第四勘察设计院集团有限公司 一种地质勘探岩芯的识别方法、装置和存储介质
CN111007064A (zh) * 2019-12-13 2020-04-14 常州大学 一种基于图像识别的录井岩性智能识别方法
WO2022031839A1 (fr) * 2020-08-07 2022-02-10 Bly Ip Inc. Systèmes et procédés d'analyse améliorée de carotte
AU2021212137B2 (en) * 2020-08-07 2022-12-01 Boart Longyear Company Systems and methods for improved core sample analysis
US11790506B2 (en) 2020-08-07 2023-10-17 Longyear Tm, Inc. Systems and methods for improved core sample analysis
CN113138178A (zh) * 2021-04-15 2021-07-20 上海海关工业品与原材料检测技术中心 一种识别进口铁矿石品牌的方法
CN113216946A (zh) * 2021-05-11 2021-08-06 西南石油大学 一种基于岩屑和元素录井数据的井壁崩落、掉块检测方法
WO2023132935A1 (fr) * 2022-01-07 2023-07-13 Schlumberger Technology Corporation Systèmes et procédés permettant de segmenter des instances de particules rocheuses
WO2023209270A1 (fr) * 2022-04-30 2023-11-02 Lumo Analytics Oy Procédé et système d'analyse de la teneur en éléments de carottes de forage
CN115410049A (zh) * 2022-10-31 2022-11-29 中国石油大学(华东) 一种岩体溶蚀程度的分类评估方法及装置
CN115410049B (zh) * 2022-10-31 2023-01-31 中国石油大学(华东) 一种岩体溶蚀程度的分类评估方法及装置

Similar Documents

Publication Publication Date Title
WO2019167030A1 (fr) Identification et enregistrement de propriétés d'échantillons de carotte
CA3019124C (fr) Description de carotte automatisee
US11530998B2 (en) Method and system to analyze geologic formation properties
CN113137954B (zh) 根据视频和地面内传感器数据生成实时传感器地图
EP3896496A1 (fr) Imagerie de déblais pour déterminer des propriétés géologiques
Bastin et al. Open-source mapping and services for Web-based land-cover validation
US11670073B2 (en) System and method for detection of carbonate core features from core images
TW201906038A (zh) 半導體晶圓檢測系統及其方法
WO2023132935A1 (fr) Systèmes et procédés permettant de segmenter des instances de particules rocheuses
CN115035297A (zh) 一种钻探岩芯rqd自动编录方法、系统、装置及介质
US20230175384A1 (en) Classification of pore or grain types in formation samples from a subterranean formation
CN108228988A (zh) 一种基于大数据驱动的边坡位移预测及滑动判别方法
CN107316131A (zh) 一种基于图像识别的电能计量装置安装工艺质量检测系统
CN117011759A (zh) 钻爆法隧道掌子面围岩多元地质信息解析方法及系统
TWM550465U (zh) 半導體晶圓檢測系統
WO2023133512A1 (fr) Systèmes et procédés de mesure de caractéristiques lithologiques physiques en fonction de photographies étalonnées de particules rocheuses
CN110226109B (zh) 增强的地质服务表征
Günther et al. Towards a machine learning framework for drill core analysis
Poleshchenko et al. Detection of Sunflower Plants in UAV Photos
WO2023076438A1 (fr) Apprentissage adaptatif pour segmentation semantique
US20230038924A1 (en) Core model augmented reality
Liu et al. Automatic identification of rock formation interface based on borehole imaging
CN115438115A (zh) 油田综合研究大数据样本标注方法
US20230134372A1 (en) Characterization of subsurface features using image logs
US11802474B2 (en) Formation-cutting analysis system for detecting downhole problems during a drilling operation

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 19760024

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 19760024

Country of ref document: EP

Kind code of ref document: A1