Simulation device and method for intelligently monitoring and predicting rock debris migration

Abstract

The invention discloses a simulation device and method for intelligently monitoring and predicting rock debris migration, the simulation device comprises a simulation wellbore, a simulation drill string, a camera shooting assembly, a rock debris injection assembly, a drilling fluid water tank and a rock debris and drilling fluid recovery box, the simulation drill string is mounted in the simulation wellbore, and a left circumscribed cylinder and a right circumscribed cylinder are respectively arranged at two ends of the simulation wellbore; two water pumps are arranged on the drilling fluid water tank, the left circumscribed cylinder and the middle of the left half part of the simulation wellbore are communicated with the two water pumps through pipelines respectively, the right circumscribed cylinder and the middle of the right half part of the simulation wellbore are communicated with the rock debris and drilling fluid recycling box through pipelines respectively, and the drilling fluid water tank and the rock debris and drilling fluid recycling box are communicated through pipelines; the rock debris injection assembly is installed on the simulation wellbore, and the camera shooting assembly shoots the simulation wellbore; and a driving assembly for driving the simulation drill column to rotate is arranged in the right circumscribed cylinder. According to the invention, most conditions encountered in the drilling process can be relatively completely simulated, and the influence of the conditions on rock debris migration is analyzed.

Description

Technical Field

[0001] The invention relates to an intelligent device and method for monitoring and predicting rock cuttings migration simulation, belonging to the technical field of oil drilling. Background Art

[0002] A major challenge faced during horizontal well drilling is that when cuttings are returned through the horizontal annulus, they easily accumulate at the bottom of the annulus to form a bed, resulting in an unclean wellbore. This brings many engineering and technical challenges, including: drill bit wear and tear, reduced drilling speed, increased wellbore instability, drill tool sticking, and increased formation contamination.

[0003] These engineering challenges severely hamper the effective development of oil and gas resources. Therefore, during drilling operations, especially in highly deviated or horizontal wells, it is crucial to accurately monitor the formation of cuttings beds, determine their thickness distribution, and promptly develop effective engineering solutions and safety measures to remove cuttings beds in risky sections. Existing technologies are unable to simulate the complex conditions of actual wellbore kicks and lost circulation, which are common in drilling. Existing technologies primarily simulate the cuttings settling process in horizontal wells, but are unable to simulate cuttings settling at varying inclinations. Summary of the Invention

[0004] The purpose of the present invention is to provide an intelligent monitoring and prediction simulation device and method for rock debris migration in response to the problems existing in the prior art.

[0005] The technical solution provided by the present invention to solve the above technical problems is: an intelligent monitoring and prediction simulation device for cuttings migration, including a simulated wellbore, a simulated drill string, a camera assembly, a cuttings injection assembly, a drilling fluid tank, cuttings and drilling fluid recovery box, the simulated drill string is installed in the simulated wellbore, and a left circumscribed cylinder and a right circumscribed cylinder are respectively provided at both ends of the simulated wellbore; two water pumps are provided on the drilling fluid tank, the left circumscribed cylinder and the middle of the left half of the simulated wellbore are connected to the two water pumps through pipes respectively, the right circumscribed cylinder and the middle of the right half of the simulated wellbore are respectively connected to the cuttings and drilling fluid recovery box through pipes, and the drilling fluid tank, cuttings and drilling fluid recovery box are connected through pipes; the cuttings injection assembly is installed on the simulated wellbore, and the camera assembly takes pictures of the simulated wellbore; a driving assembly for driving the simulated drill string to rotate is provided in the right circumscribed cylinder.

[0006] A further technical solution is that two eccentric rotor motors are provided in the simulated drill string.

[0007] A further technical solution is that the device also includes a slope adjustment component, which includes two telescopic rods and a support rod. The two ends of the simulated wellbore are respectively installed on the top of the two telescopic rods, and the middle part is installed on the top of the support rod.

[0008] A further technical solution is that a cuttings filter is provided in the cuttings and drilling fluid recovery box.

[0009] A further technical solution is that the cuttings injection assembly includes a screw feeder and a funnel, the funnel is installed on the left end of the screw feeder, and the right end of the screw feeder is connected to the left end of the simulated wellbore.

[0010] A further technical solution is that a feed valve is provided at the bottom of the funnel.

[0011] A further technical solution is that the camera assembly includes a camera, a camera bracket, and a computer, the camera is installed on the camera bracket, the computer is connected to the camera, and the camera takes pictures of the simulated wellbore.

[0012] A further technical solution is that a pressure gauge I and a flow meter I are provided between the drilling fluid tank and the left circumscribed cylinder, and a pressure gauge II and a flow meter II are provided between the simulated wellbore and the drilling fluid tank.

[0013] A visual cuttings transport simulation method comprises the following steps:

[0014] Step S1: injecting drilling fluid from the drilling fluid tank into the left circumscribed cylinder and the middle of the left half of the simulated wellbore through a water pump, and then the drilling fluid enters the simulated wellbore and the simulated drill string respectively;

[0015] Step S2: start the cuttings injection assembly to inject cuttings into the simulated wellbore;

[0016] Step S3: starting the driving assembly to drive the simulated drill string to rotate in the simulated wellbore;

[0017] Step S4: Drilling fluid and cuttings are continuously injected into the simulated wellbore, wherein a portion of the drilling fluid and cuttings directly enters the cuttings and drilling fluid recovery box in the middle of the right half of the simulated wellbore, and the other portion enters the right circumscribed cylinder and then enters the cuttings and drilling fluid recovery box;

[0018] Step S5: While continuously injecting drilling fluid and cuttings, the distribution of the cuttings bed is generated in real time by the camera assembly; the distribution of the cuttings bed generated in real time is combined with a machine learning algorithm to predict the distribution of the entire section of cuttings under the current working conditions;

[0019] Step S6: Simultaneously adjust the amplitudes of the two eccentric rotor motors in the simulated drill string to obtain the movement trajectory of the cuttings, the bed formation conditions, and the bed formation positions under different amplitudes, and analyze the effect of the drill string vibration on the rock carrying effect;

[0020] Step S7: Simultaneously adjust the telescopic heights of the two telescopic rods to simulate the wellbore inclination, obtain the movement trajectory of the rock cuttings under different wellbore inclinations, and obtain the bed formation conditions and bed formation positions, and analyze the influence of the wellbore inclination on the rock carrying effect.

[0021] A further technical solution is that the specific process of step S50 is:

[0022] Step S51, real-time image acquisition: ensuring that the two cameras synchronously acquire the distribution of the cuttings bed through software synchronization;

[0023] Step S52: pre-process the real-time image: grayscale, denoising, contrast enhancement, and improve feature extraction stability;

[0024] Step S53, stereo matching: by finding the pixel position difference of the same position point in the two camera images, calculate the three-dimensional depth of the point;

[0025] Step S54, from depth map to point cloud: for each pixel, according to the depth value and camera intrinsic parameters, obtain the three-dimensional coordinates through inverse projection transformation to generate a point cloud;

[0026] Step S55, mapping point cloud to voxel: For each point cloud point, directly map it to each voxel grid, calculate the pixel grayscale value of the point cloud in the voxel grid, average the grayscale value of each point cloud to obtain the grayscale value of the voxel, and reconstruct the distribution of the cuttings bed through the grayscale value of each voxel;

[0027] Step S56: The nine input data of the above experimental process, namely, the drill string rotation speed, drill string amplitude, drilling fluid flow rate, cuttings density, grayscale value of each voxel, volume of each cuttings section, wellbore inclination, well leakage flow rate, and well kick flow rate, are used to learn the experimental data of the previous section through a machine learning algorithm to predict the grayscale value of the voxel grid of the subsequent section, that is, to predict the volume and distribution state of the cuttings bed of the subsequent section.

[0028] Beneficial effects of the present invention:

[0029] 1. The present invention can well simulate the rock cuttings movement under drill string vibration conditions by using a built-in eccentric rotor motor;

[0030] 2. The present invention can simulate the cuttings migration condition of an inclined wellbore under real working conditions by changing the inclination of the wellbore;

[0031] 3. The present invention uses a drilling fluid loss pipe and a drilling fluid injection pipe to simulate the drilling fluid loss and well kick state under real working conditions;

[0032] 4. The present invention can fully simulate most conditions encountered during drilling and analyze the impact of these conditions on cuttings migration;

[0033] 5. The present invention can monitor the movement of rock cuttings in real time and predict the location of rock cuttings beds through machine learning algorithms;

[0034] 6. The present invention can predict the trajectory and bed position of a small amount of cuttings through the Lagrangian particle tracking method, monitor the distribution of cuttings in a certain section of the wellbore, and predict the subsequent distribution of cuttings through a machine learning algorithm. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 It is a structural schematic diagram of the present invention;

[0036] Figure 2 Schematic diagram of image processing for computer voxel gridding;

[0037] Figure 3 The rock debris distribution map taken by the horizontally positioned camera;

[0038] Figure 4 This is a computer gridded grayscale rock debris distribution map (horizontal);

[0039] Figure 5 The rock debris distribution map taken by the vertically positioned camera;

[0040] Figure 6 This is a computer gridded grayscale rock debris distribution map (vertical);

[0041] Figure 7 The three-dimensional distribution map of rock cuttings is inverted by orthogonal cross-image. DETAILED DESCRIPTION

[0042] The technical solution of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the embodiments described are only part of the embodiments of the invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making any creative efforts are within the scope of protection of the present invention.

[0043] like Figure 1As shown, an intelligent monitoring and prediction simulation device for rock cuttings migration of the present invention includes a simulated wellbore 6, a simulated drill string 8, a camera assembly, a rock cuttings injection assembly, a drilling fluid tank 19, and a rock cuttings and drilling fluid recovery tank 25. The simulated drill string 8 is installed in the simulated wellbore 6, and a left circumscribed cylinder 2 and a right circumscribed cylinder are respectively provided at both ends of the simulated wellbore 6; two water pumps 18 are provided on the drilling fluid tank 19, and the left circumscribed cylinder 2 and the middle of the left half of the simulated wellbore 6 are connected to the two water pumps 18 through pipes respectively, and the right circumscribed cylinder and the middle of the right half of the simulated wellbore 6 are connected to the rock cuttings and drilling fluid recovery tank 25 through pipes respectively, and the drilling fluid tank 19 and the rock cuttings and drilling fluid recovery tank 25 are connected through pipes; the rock cuttings injection assembly is installed on the simulated wellbore 6, and the camera assembly takes pictures of the simulated wellbore 6; a driving assembly for driving the simulated drill string 8 to rotate is provided in the right circumscribed cylinder. A pressure gauge I 16 and a flow meter I 17 are provided between the drilling fluid tank 19 and the left external cylinder 2, a pressure gauge II 21 and a flow meter II 20 are provided between the simulated wellbore 6 and the drilling fluid tank 19; a well leakage flow meter 23 is provided between the simulated wellbore 6 and the rock cuttings and drilling fluid recovery tank 25, and a reverse valve 24 is provided between the right external cylinder and the rock cuttings and drilling fluid recovery tank 25.

[0044] In this embodiment, the simulated wellbore 6 and the simulated drill string 8 are both PVC pipes, and the connecting pipes are all steel hoses 7.

[0045] The working process of the simulation device is as follows: two water pumps 18 respectively inject drilling fluid into the left circumscribed cylinder 2 and the middle of the left half of the simulated wellbore 6 for drilling fluid injection and drilling fluid kick state; at the same time, cuttings are added to the simulated wellbore 6 through the cuttings injection assembly, and the simulated drill string 8 is rotated at the same time; after continuous injection, a part of the drilling fluid and cuttings directly enter the cuttings and drilling fluid recovery box 25 in the middle of the right half of the simulated wellbore 6 (for simulating well leakage state), and the other part enters the right circumscribed cylinder and then enters the cuttings and drilling fluid recovery box 25 (for simulating drilling fluid backflow).

[0046] On the basis of this embodiment, in order to simulate and analyze the influence of drill string vibration on the rock carrying effect, a preferred implementation method is that two eccentric rotor motors 7 are provided in the simulated drill string 8, wherein the two eccentric rotor motors 7 are respectively located at the left and right ends of the simulated drill string 8. The eccentric rotor motors 7 are built into a PVC pipe and the amplitude is adjusted by remotely controlling the speed to achieve simulation of drill string vibration.

[0047] On the basis of this embodiment, in order to simulate the inclination of the wellbore, a preferred implementation method is that the device also includes an inclination adjustment component, which includes two telescopic rods 27 and a support rod 9. The two ends of the simulated wellbore 6 are respectively installed on the top of the two telescopic rods 27, and the middle part is installed on the top of the support rod 9. That is, by adjusting the telescopic height of the two telescopic rods 27, the inclination of the wellbore can be simulated, thereby analyzing the influence of the wellbore inclination on the rock carrying effect.

[0048] On the basis of this embodiment, in order to collect cuttings more conveniently, a preferred implementation manner is that a cuttings filter 26 is provided in the cuttings and drilling fluid recovery box 25 .

[0049] In this embodiment, the cuttings injection assembly includes a screw feeder 5 and a funnel 3. The funnel 3 is installed on the left end of the screw feeder 5. The right end of the screw feeder 5 is connected to the left end of the simulated wellbore 6. A feed valve 4 is provided at the bottom of the funnel 3 for controlling the feed amount of cuttings.

[0050] In actual use, the cuttings are added to the funnel 3 in advance, and then the feeding amount of the cuttings is controlled by switching the feeding valve 4.

[0051] In this embodiment, the camera assembly includes a camera 12 , a camera bracket 13 , and a computer 14 . The camera 12 is mounted on the camera bracket 13 . The computer 14 is connected to the camera 12 . The camera 12 takes pictures of the simulated wellbore 6 .

[0052] A visual cuttings transport simulation method comprises the following steps:

[0053] Step S1: The drilling fluid in the drilling fluid tank 19 is injected into the left circumscribed cylinder 2 and the middle of the left half of the simulated wellbore 6 by the water pump 18, and then the drilling fluid enters the simulated wellbore 6 and the simulated drill string 8 respectively;

[0054] Step S2: start the cuttings injection assembly to inject cuttings into the simulated wellbore 6;

[0055] Step S3: starting the driving assembly, which drives the simulated drill string 8 to rotate in the simulated wellbore 6;

[0056] Step S4: Drilling fluid and cuttings are continuously injected into the simulated wellbore 6, wherein a portion of the drilling fluid and cuttings directly enters the cuttings and drilling fluid recovery tank 25 in the middle of the right half of the simulated wellbore 6, and the other portion enters the right circumscribed cylinder and then enters the cuttings and drilling fluid recovery tank 25;

[0057] Step S5: While continuously injecting drilling fluid and cuttings, the distribution of the cuttings bed is generated in real time by the camera assembly; the distribution of the cuttings bed generated in real time is combined with a machine learning algorithm to predict the distribution of the entire section of cuttings under the current working conditions;

[0058] In step S5, two sets of vertically positioned cameras are used to record the generation and movement of rock cuttings in real time. The two cameras are fixed on two perpendicular axes, ensuring that the optical axes are perpendicular and the overlapping field of view covers the rock cuttings bed. The two vertically positioned cameras (with orthogonal optical axes) acquire images of the rock cuttings bed from different perspectives, calculate pixel depth using the parallax principle, and voxelize the 3D structure, assigning different grayscale values ​​to each voxel to reconstruct the 3D structure.

[0059] The specific steps include:

[0060] Step S51, real-time image acquisition: ensure that the two cameras synchronously acquire the distribution of the cuttings bed through software synchronization (time stamp alignment);

[0061] Step S52: pre-process the real-time image: grayscale, denoising (Gaussian filtering), contrast enhancement, and improve feature extraction stability;

[0062] Step S53, stereo matching: by finding the pixel position difference (parallax) of the same position point in the two camera images, calculate the 3D depth of the point;

[0063] Step S54, from depth map to point cloud: for each pixel (u, v), according to the depth value Z and the camera intrinsic parameters, the three-dimensional coordinates (X, Y, Z) are obtained by inverse projection transformation, and the point cloud P = {X i , Y i , Z i};

[0064] Step S55, mapping point cloud to voxel: For each point cloud point, directly map it to each voxel grid, calculate the pixel grayscale value of the point cloud in the voxel grid, average the grayscale value of each point cloud to obtain the grayscale value of the voxel, and reconstruct the distribution of the cuttings bed through the grayscale value of each voxel;

[0065] Step S56: Using the nine input data of the above experimental process, namely, the drill string rotation speed, drill string amplitude, drilling fluid flow rate, cuttings density, grayscale value of each voxel, volume of each cuttings section, wellbore inclination, lost circulation flow rate, and kick flow rate, a machine learning algorithm is used to learn the experimental data of the first section to predict the grayscale value of the voxel grid of the latter section, that is, to predict the volume and distribution of the cuttings bed of the latter section;

[0066] Step S6: Simultaneously adjust the amplitudes of the two eccentric rotor motors 7 in the simulated drill string 8 to obtain the movement trajectory of the rock cuttings, the bed formation conditions, and the bed formation positions under different amplitudes, and analyze the influence of the drill string vibration on the rock carrying effect;

[0067] Step S7: Simultaneously adjust the telescopic heights of the two telescopic rods 27 to simulate the wellbore inclination, obtain the movement trajectory of the rock cuttings under different wellbore inclinations, and the bed formation conditions and positions, and analyze the influence of the wellbore inclination on the rock carrying effect.

[0068] In this embodiment, different experimental devices can also be adjusted to analyze the influence of different factors on the rock carrying effect;

[0069] Different amounts of cuttings were injected through the cuttings injection assembly, and the movement trajectory of the cuttings, the bed formation and the bed position of the cuttings were obtained respectively under different cuttings amounts, and the influence of different cuttings amounts on the rock carrying effect was analyzed;

[0070] By adjusting the rotation speed of the water pump 18, the displacement of the drilling fluid is adjusted; the movement trajectory of the cuttings, the bed formation and the bed position of the cuttings under different drilling fluid displacements are obtained, and the influence of different drilling fluid flow rates on the rock carrying effect is analyzed;

[0071] Adjust the loss flow rate under the cuttings loss state by valve, and analyze the influence of the loss flow rate under the cuttings loss state on the rock carrying effect;

[0072] By adjusting the rotation speed of the simulated drill string 8 through the drive assembly, the movement trajectory of the rock cuttings, the bed formation and bed position of the rock cuttings at different drill string rotation speeds are obtained, and the influence of different drill rod rotation speeds on the rock carrying effect is analyzed.

[0073] The above description does not limit the present invention in any form. Although the present invention has been disclosed through the above embodiments, it is not intended to limit the present invention. Any technician familiar with the profession can use the technical content disclosed above to make some changes or modifications to equivalent embodiments without departing from the scope of the technical solution of the present invention. However, any simple modifications, equivalent changes and modifications made to the above embodiments based on the technical essence of the present invention without departing from the content of the technical solution of the present invention are within the scope of the technical solution of the present invention.

Claims

1. An intelligent monitoring and prediction device for rock debris migration, characterized in that: The invention comprises a simulated wellbore (6), a simulated drill string (8), a camera assembly, a cuttings injection assembly, a drilling fluid water tank (19), and a cuttings and drilling fluid recovery tank (25). The simulated drill string (8) is installed in the simulated wellbore (6). The two ends of the simulated wellbore (6) are respectively provided with a left circumscribed cylinder (2) and a right circumscribed cylinder; the drilling fluid water tank (19) is provided with two water pumps (18); the left circumscribed cylinder (2) and the middle of the left half of the simulated wellbore (6) are respectively connected through pipes. The right circumscribed cylinder and the middle of the right half of the simulated wellbore (6) are respectively connected to the rock cuttings and drilling fluid recovery box (25) through pipelines, and the drilling fluid water tank (19), rock cuttings and drilling fluid recovery box (25) are connected through pipelines; the rock cuttings injection assembly is installed on the simulated wellbore (6), and the camera assembly takes a picture of the simulated wellbore (6); a driving assembly for driving the simulated drill string (8) to rotate is provided in the right circumscribed cylinder.

2. The intelligent monitoring and prediction simulation device for cuttings migration according to claim 1, characterized in that: Two eccentric rotor motors (7) are provided in the simulated drill string (8).

3. The intelligent monitoring and prediction simulation device for rock debris migration according to claim 1, characterized in that: The device also includes an inclination adjustment component, which includes two telescopic rods (27) and a support rod (9). The two ends of the simulated wellbore (6) are respectively installed on the top ends of the two telescopic rods (27), and the middle part is installed on the top end of the support rod (9).

4. The intelligent monitoring and prediction simulation device for rock debris migration according to claim 1, characterized in that: A cuttings filter (26) is provided in the cuttings and drilling fluid recovery box (25).

5. The intelligent monitoring and prediction simulation device for rock debris migration according to claim 1, characterized in that: The cuttings injection assembly comprises a screw feeder (5) and a funnel (3), wherein the funnel (3) is mounted on the left end of the screw feeder (5), and the right end of the screw feeder (5) is connected to the left end of the simulated wellbore (6).

6. The intelligent monitoring and prediction simulation device for rock debris migration according to claim 5, characterized in that: A feed valve (4) is provided at the bottom of the funnel (3).

7. The intelligent monitoring and prediction simulation device for rock debris migration according to claim 1, characterized in that: The camera assembly comprises a camera (12), a camera bracket (13), and a computer (14); the camera (12) is mounted on the camera bracket (13); the computer (14) is connected to the camera (12); and the camera (12) takes pictures of the simulated wellbore (6).

8. The intelligent monitoring and prediction simulation device for rock debris migration according to claim 1, characterized in that: A pressure gauge I (16) and a flow meter I (17) are provided between the drilling fluid tank (19) and the left circumscribed cylinder (2), and a pressure gauge II (21) and a flow meter II (20) are provided between the simulated wellbore (6) and the drilling fluid tank (19).

9. A visual cuttings transport simulation method, characterized in that: The method uses an intelligent monitoring and prediction rock debris migration simulation device according to any one of claims 1 to 8 to perform simulation, and specifically comprises the following steps: Step S1: The drilling fluid in the drilling fluid tank (19) is injected into the left circumscribed cylinder (2) and the middle of the left half of the simulated wellbore (6) through the water pump (18), and then the drilling fluid enters the simulated wellbore (6) and the simulated drill string (8) respectively; Step S2, starting the cuttings injection assembly to inject cuttings into the simulated wellbore (6); Step S3: starting the driving assembly, which drives the simulated drill string (8) to rotate in the simulated wellbore (6); Step S4, continuously injecting drilling fluid and rock cuttings into the simulated wellbore (6), wherein a portion of the drilling fluid and rock cuttings directly enters the rock cutting and drilling fluid recovery box (25) in the middle of the right half of the simulated wellbore (6), and the other portion enters the right circumscribed cylinder and then enters the rock cutting and drilling fluid recovery box (25); Step S5: While continuously injecting drilling fluid and cuttings, the distribution of the cuttings bed is generated in real time by the camera assembly; the distribution of the cuttings bed generated in real time is combined with a machine learning algorithm to predict the distribution of the entire section of cuttings under the current working conditions; Step S6: Simultaneously adjust the amplitudes of the two eccentric rotor motors (7) in the simulated drill string (8), obtain the movement trajectory of the rock cuttings under different amplitudes, the bed formation conditions and the bed formation positions, and analyze the influence of the drill string vibration on the rock carrying effect; Step S7, simultaneously adjusting the telescopic heights of the two telescopic rods (27), simulating the wellbore inclination, obtaining the movement trajectory of the rock cuttings under different wellbore inclinations, the bed formation conditions and the bed formation positions, and analyzing the influence of the wellbore inclination on the rock carrying effect.

10. A visual cuttings transport simulation method according to claim 1, characterized in that: The specific process of step S5 is as follows: Step S51, real-time image acquisition: ensuring that the two cameras synchronously acquire the distribution of the cuttings bed through software synchronization; Step S52: pre-process the real-time image: grayscale, denoising, contrast enhancement, and improve feature extraction stability; Step S53, stereo matching: by finding the pixel position difference of the same position point in the two camera images, calculate the three-dimensional depth of the point; Step S54, from depth map to point cloud: for each pixel, according to the depth value and camera intrinsic parameters, obtain the three-dimensional coordinates through inverse projection transformation to generate a point cloud; Step S55, mapping point cloud to voxel: For each point cloud point, directly map it to each voxel grid, calculate the pixel grayscale value of the point cloud in the voxel grid, average the grayscale value of each point cloud to obtain the grayscale value of the voxel, and reconstruct the distribution of the cuttings bed through the grayscale value of each voxel; Step S56: The nine input data of the above experimental process, namely, the drill string rotation speed, drill string amplitude, drilling fluid flow rate, cuttings density, grayscale value of each voxel, volume of each cuttings section, wellbore inclination, well leakage flow rate, and well kick flow rate, are used to learn the experimental data of the previous section through a machine learning algorithm to predict the grayscale value of the voxel grid of the subsequent section, that is, to predict the volume and distribution state of the cuttings bed of the subsequent section.

Images