Method and system for construction of artificial intelligence model using on-cutter sensing data for predicting well bit performance

1. A system comprising:

an instrumented cutter of a drill bit comprising an on-cutter sensor for monitoring drilling performance metrics while performing drilling operations based on offset well data; and

a computing device comprising a model development system configured to use the drilling performance metrics, surface drilling parameters, and characteristics of the instrumented cutter to train a machine learning (ML) model,

wherein the trained ML model is used to optimize drilling parameters and predict drill bit performance in a current well,

wherein the on-cutter sensor is at least one ultrasonic transducer for measuring wear of a polycrystalline diamond table, by exciting the polycrystalline diamond table with an ultrasonic pulse and recording an ultrasonic vibration of the polycrystalline diamond table.

2. The system of claim 1 , wherein the ML model is a deep learning (DL) machine learning model.

3. The system of claim 1 , wherein the instrumented cutter comprises a plurality of on-cutter sensors for monitoring cutter wear.

4. The system of claim 3 , wherein the computing device further comprises a data processing module configured to aggregate the drilling performance metrics measured by the plurality of on-cutter sensors and transform the processed data to remove noise.

5. The system of claim 1 , wherein the drilling performance metrics comprise at least one selected from the group consisting of a strain, an acceleration, a motion, a vibration, an image, an electrical resistance, a temperature, a pressure, an electrical capacitance, an electrical inductance, a magnetic field, and a photoelectric emission.

6. The system of claim 1 , wherein the instrumented cutter comprises a wireless transmitter configured to transmit the drilling performance parameters to the computing device.

7. The system of claim 1 , wherein surface drilling parameters that are optimized using the ML model comprise a weight on bit, a rotational speed, a torque on bit, a downhole mud pressure, and a downhole mud flow rate.

8. A method, comprising:

measuring drilling performance metrics while performing drilling operations in an offset well using an instrumented cutter of a drill bit comprising an on-cutter sensor;

transmitting the drilling performance metrics to a computing device;

training a machine learning (ML) model hosted by the computing device using processed drilling performance metrics, surface drilling parameters, and characteristics of the instrumented cutter to obtain a trained ML model;

using the trained ML model to optimize the surface drilling parameters and predict drill bit performance while performing drilling operations in a current well;

processing the drill performance metrics, surface drilling parameters, and the characteristics of the instrumented cutter by:

aggregating drill performance metrics from a plurality of on-cutter sensors of the instrumented cutter; and

transforming the drill performance metrics, the surface drilling parameters, and characteristics of the instrumented cutter to remove noise using a Fast Fourier Transform.

9. The method of claim 8 , wherein the ML model is a deep learning (DL) machine learning model.

10. The method of claim 8 , wherein surface drilling parameters that are optimized using the ML model comprise a weight on bit, a rotational speed, a torque on bit, a downhole mud pressure, and a downhole mud flow rate.

11. The method of claim 8 , further comprising: using the trained ML model to predict drill bit performance on a new well in geographical proximity to the offset well.

12. The method of claim 8 , wherein the instrumented cutter comprises a wireless transmitter configured to wirelessly transmit the drilling performance parameters to the computing device.

13. The method of claim 8 , wherein the drilling performance metrics measured comprise at least one selected from the group consisting of a strain, an acceleration, a motion, a vibration, an image, an electrical resistance, a temperature, a pressure, an electrical capacitance, an electrical inductance, a magnetic field, and a photoelectric emission.

14. A non-transitory computer readable medium comprising software instructions, that when executed, perform a method comprising:

measuring drilling performance metrics while performing drilling operations in an offset well using an instrumented cutter of a drill bit comprising an on-cutter sensor;

transmitting the drilling performance metrics to a computing device;

training a machine learning (ML) model hosted by the computing device using processed drilling performance metrics, surface drilling parameters, and characteristics of the instrumented cutter to obtain a trained ML model;

using the trained ML model to optimize the surface drilling parameters and predict drill bit performance while performing drilling operations in the offset well, and

processing the drill performance metrics, surface drilling parameters, and the characteristics of the instrumented cutter by:

aggregating drill performance metrics from a plurality of on-cutter sensors of the instrumented cutter; and

transforming the drill performance metrics, the surface drilling parameters, and characteristics of the instrumented cutter to remove noise using a Fast Fourier Transform.