Granted Patent B2

US 12,378,878 · App. 17/708,379 · Granted Aug 5, 2025

Inventors: Ahmed I. Saihati (Saihat, SA); Bader M. Otaibi (Dhahran, SA)

Assignee: SAUDI ARABIAN OIL COMPANY

E21B49/003E21B44/00E21B45/00E21B2200/20

Abstract

Claims (87)

1. A method, comprising:

obtaining first drilling surface parameter data regarding a plurality of drilling parameters during a first drilling operation for a first wellbore, wherein the first drilling surface parameter data comprises drilling rotary speed data, first mud pump rate data, and drilling weight-on-bit data that are being used in the first drilling operation;

obtaining first geological data regarding a geological formation within a subsurface of the first wellbore that is being drilled during the first drilling operation;

obtaining first historical vibration data regarding one or more drilling operations for one or more wellbores;

obtaining loss event data regarding one or more lost circulation events from the first wellbore;

determining, by a computer processor, first predicted vibration data for a bottomhole assembly in the first drilling operation using a first machine-learning model, the first drilling surface parameter data, the first geological data, the first historical vibration data, and a first rate of penetration (ROP) value regarding the bottomhole assembly,

wherein the first predicted vibration data comprises predicted lateral vibration data and predicted torsional vibration data,

wherein the first machine-learning model is a first artificial neural network comprising a first input layer, a first plurality of hidden layers, and a first output layer, and

wherein the first drilling surface parameter data, the loss event data, the first geological data, and the first historical vibration data are inputs to the first input layer of the first artificial neural network, and

wherein the predicted lateral vibration data and the predicted torsional vibration data are outputs from the first output layer of the first artificial neural network;

determining, by the computer processor, a predicted ROP value regarding the bottomhole assembly using the first predicted vibration data, a second machine-learning model, and the first ROP value,

wherein the second machine-learning model is a second artificial neural network comprising a second input layer, a second plurality of hidden layers, and a second output layer,

wherein the first predicted vibration data, the first drilling surface parameter data, and the first ROP value are inputs to the second input layer of the second artificial neural network, and

wherein the predicted ROP value is an output from the second output layer of the second artificial neural network;

determining, by the computer processor, a plurality of drilling parameter clusters using the predicted ROP value, the first predicted vibration data, a plurality of drilling parameter combinations, and a clustering algorithm,

wherein the plurality of drilling parameter combinations comprises the first drilling surface parameter data,

wherein the first drilling surface parameter data corresponds to a current combination of drilling parameters being used in the first drilling operation;

presenting, by the computer processor, the plurality of drilling parameter clusters in a graphical user interface on a user device;

determining, by the computer processor, an adjusted mud pump rate based on a selected cluster among the plurality of drilling parameter clusters,

wherein the selected cluster corresponds to a user selection within the graphical user interface in response to presenting the plurality of drilling parameter clusters; and

transmitting a first command to a mud pump system to update the first drilling operation based on the adjusted mud pump rate,

wherein the mud pump system supplies a first drilling fluid to the first wellbore at the adjusted mud pump rate.

2. The method of claim 1 ,

wherein the second machine-learning model determines the predicted ROP value based on a plurality of inputs for a first section of a wellbore in the first drilling operation,

wherein the plurality of inputs comprise a second ROP value, and

wherein the second ROP value corresponds to a second section of the wellbore that was drilled prior to drilling the first section of the wellbore.

3. The method of claim 1 ,

wherein the first historical vibration data corresponds to a vibration type selected from a group consisting of a lateral vibration, a torsional vibration, and an axial vibration of a bottomhole assembly.

4. The method of claim 1 ,

wherein the first machine-learning model uses second predicted vibration data that is determined by the first machine-learning model at an earlier time than the first predicted vibration data in the first drilling operation to determine the first predicted vibration data.

5. The method of claim 1 , further comprising:

acquiring the first historical vibration data from a second wellbore using a plurality of downhole pressure sensors coupled to a drill string,

wherein the first drilling operation is performed in the first wellbore using the bottomhole assembly that does not include a downhole pressure sensor for detecting vibrations.

6. The method of claim 1 , further comprising:

obtaining a training dataset comprising second drilling surface parameter data, second geological data, second vibration data, and ROP data from a plurality of drilling operations for a plurality of wells;

obtaining an initial model; and

updating the initial model using the training dataset and a plurality of machine-learning epochs to produce a trained model,

wherein the trained model is the first machine-learning model.

7. The method of claim 1 , further comprising:

obtaining, by a user device, the first predicted vibration data of the bottomhole assembly;

presenting, on a display device coupled to the user device, a plurality of adjusted ROP values associated with the first predicted vibration data;

obtaining, by the user device, a user selection of the plurality of adjusted ROP values; and

the transmitting a second command to produce an adjusted ROP value correspond to the user selection.

8. A system, comprising:

a first drilling system comprising a bottomhole assembly that comprises a first drill string, wherein the first drilling system is coupled to a first wellbore;

a mud pump system coupled to the first wellbore; and

a control system coupled to the first drilling system and the mud pump system, wherein the control system comprises a computer processor, the control system being configured to perform a method comprising:

obtaining first drilling surface parameter data regarding a plurality of drilling parameters during a first drilling operation for the first wellbore, wherein the first drilling surface parameter data comprises drilling rotary speed data, first mud pump rate data from the mud pump system, and drilling weight-on-bit data that are being used in the first drilling operation;

obtaining first geological data regarding a geological formation within a subsurface of the first wellbore that is being drilled during the first drilling operation;

obtaining first historical vibration data regarding one or more drilling operations for one or more wellbores;

obtaining loss event data regarding one or more lost circulation events from the first wellbore;

determining first predicted vibration data for the bottomhole assembly in the first drilling operation using a first machine-learning model, the first drilling surface parameter data, the geological formation being drilled, the first historical vibration data, and a first rate of penetration (ROP) value regarding the bottomhole assembly,

wherein the first predicted vibration data comprises predicted lateral vibration data and predicted torsional vibration data,

wherein the first machine-learning model is a first artificial neural network comprising a first input layer, a first plurality of hidden layers, and a first output layer, and

wherein the first drilling surface parameter data, the loss event data, the first geological data, and the first historical vibration data are inputs to the first input layer of the first artificial neural network, and

wherein the predicted lateral vibration data and the predicted torsional vibration data are outputs from the first output layer of the first artificial neural network;

determining a predicted ROP value regarding the bottomhole assembly using the first predicted vibration data, a second machine-learning model, and the first ROP value,

wherein the second machine-learning model is a second artificial neural network comprising a second input layer, a second plurality of hidden layers, and a second output layer,

wherein the first predicted vibration data, the first drilling surface parameter data, and the first ROP value are inputs to the second input layer of the second artificial neural network, and

wherein the predicted ROP value is an output from the second output layer of the second artificial neural network;

determining a plurality of drilling parameter clusters using the predicted ROP value, the first predicted vibration data, a plurality of drilling parameter combinations, and a clustering algorithm,

wherein the plurality of drilling parameter combinations comprises the first drilling surface parameter data,

wherein the first drilling surface parameter data corresponds to a current combination of drilling parameters being used in the first drilling operation;

presenting the plurality of drilling parameter clusters in a graphical user interface on a user device;

determining an adjusted mud pump rate based on a selected cluster among the plurality of drilling parameter clusters,

wherein the selected cluster corresponds to a user selection within the graphical user interface in response to presenting the plurality of drilling parameter clusters; and

transmitting a first command to the mud pump system to update the first drilling operation based on the adjusted mud pump rate,

wherein the mud pump system is configured to supply a first drilling fluid to the first wellbore at the adjusted mud pump rate.

9. The system of claim 8 , further comprising:

a user device coupled to the control system,

wherein the user device is configured to provide the graphical user interface for presenting a plurality of predicted ROP values for a drilling operation,

wherein the method further comprises transmitting a second command to produce an adjusted ROP value, and

wherein the adjusted ROP value corresponds to a user selection that is obtained from a user using the user device.

10. The system of claim 8 ,

wherein the second machine-learning model determines the predicted ROP value based on a plurality of inputs for a first section of a wellbore in the first drilling operation,

wherein the plurality of inputs comprise a second ROP value, and

wherein the second ROP value corresponds to a second section of the first wellbore that was drilling prior to drilling the first section of the first wellbore.

11. The system of claim 8 ,

wherein the first machine-learning model uses second predicted vibration data that is determined by the first machine-learning model at an earlier time than the first predicted vibration data in the first drilling operation to determine the first predicted vibration data.

12. The system of claim 8 ,

wherein the first historical vibration data is acquired from a second wellbore using a plurality of downhole pressure sensors coupled to a second drilling system that is separate from the first drilling system, and

wherein the first drilling operation is performed in the first wellbore using the bottomhole assembly that does not include a downhole pressure sensor for detecting vibrations of the first drill string.

13. The system of claim 8 , wherein the control system is further configured to:

obtain a training dataset comprising second drilling surface parameter data, second geological data, second vibration data, and ROP data from a plurality of drilling operations for a plurality of wells;

obtain an initial model; and

update the initial model using the training dataset and a plurality of machine-learning epochs to produce a trained model,

wherein the trained model is the first machine-learning model.

Assignments (1)

Continuity (1)

References Cited (25)