Analysis of drilling slurry solids by image processing

Patent metadata

Field	Value
Publication number	US-11719089-B2
Application number	US-202016930014-A
Country	US
Kind code	B2
Filing date	Jul 15, 2020
Priority date	Jul 15, 2020
Publication date	Aug 8, 2023
Grant date	Aug 8, 2023

Abstract

Analysis of drilling slurry solids by image processing includes receiving images of a solid slurry including a mixture of wellbore drilling mud and solid objects found in the wellbore while drilling the wellbore through a subterranean zone. The solid objects include drill cuttings and non-drilled solids. By implementing image processing techniques on the images, the solids in the solid slurry are classified a drill cutting or as a non-drilled solid.

First claim

The invention claimed is:

1. A system comprising:
a digital imaging device mounted to a non-vibrating member of a shale shaker of a wellbore drilling assembly, the shale shaker positioned at a surface of the Earth adjacent a wellbore and configured to receive a solid slurry comprising a mixture of wellbore drilling mud and solid objects found in the wellbore while drilling the wellbore through a subterranean zone, the solid objects comprising drill cuttings and non-drilled solids, the digital imaging device oriented to face a portion of the shale shaker that receives the solid slurry, the digital imaging device configured to capture digital images of the solid objects while the solid slurry is received by the shale shaker; and
a computer system operatively coupled to the digital imaging device, the computer system comprising:
one or more processors; and
a computer-readable medium storing instructions executable by the one or more processors to perform operations comprising:
receiving the images captured by the digital imaging device; and
by implementing image processing techniques on the images, classifying a first solid object captured in a first image as a drill cutting and a second solid object captured in a second image as a non-drilled solid, wherein, to implement the image processing techniques, the computer system is configured to perform operations comprising deploying deep learning techniques comprising deploying a convolutional neural network (CNN) model for pattern recognition and image classification of the digital images captured by the digital imaging device,
wherein the operations further comprise:
receiving drilling parameters applied to the wellbore drilling assembly to drill the wellbore;
applying the received drilling parameters as inputs to the CNN model, wherein the drilling parameters comprise a rate of penetration, a weight on bit, and wellbore drilling mud flow rate; and
deploying the CNN model including the received drilling parameters as inputs for the pattern recognition and the image classification of the digital images.

2. The system of claim 1 , wherein an image of a solid object comprises a plurality of particles including particles representing the solid object in a first color and particles representing remaining portions of the image in a second color different from the first, wherein, to implement the image processing techniques, the computer system is configured to perform operations comprising:
counting a number of particles in the image;
identifying the particles representing the solid object, wherein the particles representing the solid object define an area representing dimensions of the solid object;
determining a long axis of the area;
determining a short axis of the area, the short axis being perpendicular to the long axis; and
determining the dimensions of the solid object based, in part, on a length of the long axis and a length of the short axis.

3. The system of claim 2 , wherein the computer system is configured to perform operations comprising:
storing dimensions of known drill cuttings and known non-drilled solids;
comparing the dimensions of the solid object with the stored dimensions; and
identifying the solid object as a drill cutting or a non-drilled solid based on a result of the comparing.

4. The system of claim 1 , wherein the computer system is configured to perform operations comprising:
receiving drill bit parameters of the drill bit of the wellbore drilling assembly, the drill bit parameters comprising a number of blades on the drill bit, a type of the drill bit and a size of the drill bit;
receiving wellbore drilling mud parameters of the wellbore drilling mud, the wellbore drilling mud parameters comprising mud weight, a mud type comprising an oil-based mud type or a water-based mud type, and rheological properties of the wellbore drilling mud;
applying the received drill bit parameters and the received wellbore drilling mud parameters as inputs to the CNN model; and
deploying the CNN model including the received drill bit parameters and the received wellbore drilling mud parameters as inputs for the pattern recognition and the image classification of the digital images.

5. The system of claim 1 , further comprising the shale shaker, wherein the shale shaker comprises a mesh on which the solid slurry flows, wherein the mesh comprises a wet zone, an intermediate zone and a dry zone, wherein the solid slurry flows past each of the wet zone, the intermediate zone and the dry zone, wherein the non-vibrating member is a horizontal bar attached to the dry zone of the mesh.

6. The system of claim 1 , wherein the digital imaging device is configured to capture the images over a period of time while the wellbore drilling assembly is drilling the wellbore, wherein the computer system is configured to perform operations comprising:
receiving the images captured by the digital imaging device over the period of time;
after classifying each solid object captured in an image received over the period of time as a drill cutting or as a non-drilled solid, storing the classification of the image with a time instant at which the digital imaging device captured the image; and
determining an existence of a wellbore condition based on classifications of the images captured over the period of time; and
modifying a drilling condition operating the wellbore drilling assembly in response to determining the existence of the wellbore condition.

7. The system of claim 6 , wherein the wellbore condition is stuck pipe, wherein the computer system is configured to perform operations comprising:
determining the stuck pipe responsive to identifying an increase in classification of images as non-drilled solids comprising wellbore cavings; and
modifying a wellbore drilling mud flow rate through the wellbore drilling system responsive to determining the stuck pipe.

8. The system of claim 1 , wherein the subterranean zone comprises a formation through which the wellbore is being drilled by the wellbore drilling assembly, wherein a drill cutting comprises a rock separated from the formation in response to a drill bit of the wellbore drilling assembly contacting the formation while drilling the wellbore, wherein the non-drilled solid is any solid object that is not a drill cutting.

9. The system of claim 8 , wherein the non-drilled solid comprises at least one of cavings, cement, elastomers or fish eyes.

10. A method comprising:
receiving, by one or more processors, images captured by a digital imaging device mounted to a non-vibrating member of a shale shaker of a wellbore drilling assembly, the shale shaker positioned at a surface of the Earth adjacent a wellbore and configured to receive a solid slurry comprising a mixture of wellbore drilling mud and solid objects found in the wellbore while drilling the wellbore through a subterranean zone, the solid objects comprising drill cuttings and non-drilled solids, the digital imaging device oriented to face a portion of the shale shaker that receives the solid slurry, the digital imaging device configured to capture digital images of the solid objects while the solid slurry is received by the shale shaker; and
by implementing image processing techniques on the images:
classifying, by the one or more processors, a first solid object captured in a first image as a drill cutting, and
classifying, by the one or more processors, a second solid object captured in a second image as a non-drilled solid,

wherein implementing the image processing techniques comprises deploying deep learning techniques comprising deploying a convolutional neural network (CNN) model for pattern recognition and image classification of t

Assignees

• Saudi Arabian Oil Co

Inventors

• Affleck Michael
• Magana Mora Arturo
• Pasan Gooneratne Chinthaka
• Contreras Otalvora William
• Singh Pratyush

Classifications

• E21B47/0025 Primary

  generating an image of the borehole wall using down-hole measurements, e.g. acoustic or electric · CPC title

• E21B21/066

  with further treatment of the solids, e.g. for disposal · CPC title

• E21B49/003

  by analysing drilling variables or conditions (E21B49/005 takes precedence; systems specially adapted for monitoring a plurality of drilling variables or conditions E21B44/00) · CPC title

• G06F18/24

  Classification techniques · CPC title

• G06N3/02

  Neural networks · CPC title