Granted Patent B2

US 12,554,040 · App. 18/006,602 · Granted Feb 17, 2026

Inventors: Mandar Shrikant Kulkarni (Yerawada, IN); Purnaprajna Raghavendra Mangsuli (Pune, IN); Hiren Maniar (Houston, TX); Aria Abubakar (Sugar Land, TX)

Assignee: SCHLUMBERGER TECHNOLOGY CORPORATION

G01V20/00G06N20/00

Abstract

Claims (44)

1 . A method for correlating well logs, comprising:

taking measurements along an entirety of a well or along a portion of the well via an instrument deployed into the well;

generating a well log based upon recordings of the measurements;

receiving the well log;

applying the well log as a first input to a first machine learning model comprising an encoder-decoder architecture with skip connections between mirroring layers of an encoder and a decoder, wherein the first machine learning model is configured to predict first markers in the well log based at least in part on a global factor of the well log;

applying the well log as a second input to a second machine learning model comprising a convolutional neural network having a plurality of one-dimensional convolution layers that are stacked together, wherein the second machine learning model is configured to predict second markers in the well log based at least in part on a local factor of the well log;

generating a set of predicted well markers by merging at least some of the first markers and at least some of the second markers;

aligning the well log with respect to one or more other well logs based at least in part on the set of predicted well markers;

generating a digital model of a subterranean volume through which the well log extends based at least in part on the well log that was correlated to the other well logs; and

controlling drilling parameters as part of a drilling operation based on the digital model, wherein controlling the drilling parameters comprises controlling a drilling trajectory, selecting and controlling a type of equipment utilized in the drilling operation, or controlling parameters of the type of equipment utilized in the drilling operation.

2 . The method of claim 1 , further comprising:

receiving a training set of one or more well logs having one or more markers associated with the one or more well logs, and

raining the first machine learning model and the second machine learning model using the training set.

3 . The method of claim 2 , further comprising smoothening one or more markers of the training set prior to training the first machine learning model and prior to training the second machine learning model.

4 . The method of claim 3 , wherein smoothening comprises convolving the one or more markers with a Gaussian kernel.

5 . The method of claim 1 , wherein the global factor represents a characteristic of the well log as a whole, and wherein the local factor represents a characteristic of an individual layer of a plurality of layers represented in the well log.

6 . The method of claim 1 , comprising applying the well log as the second input to the second machine learning model in parallel with applying the well log as the first input to the first machine learning model.

7 . The method of claim 1 , wherein merging at least some of the first markers and at least some of the second markers results in a reduction in error marker pickings of the first markers and a reduction in false positives from the second markers.

8 . A non-transitory computer-readable medium storing instructions that, when executed by one or more processors of a computing system, cause the computing system to perform operations, the operations comprising:

receiving a well log generated based upon recordings of measurements taken along an entirety of a well or along a portion of the well via an instrument deployed into the well;

applying the well log as a first input to a first machine learning model the comprising an encoder-decoder architecture with skip connections between mirroring layers of an encoder and a decoder, wherein the first machine learning model is configured to predict first markers in the well log based at least in part on a global factor of the well log;

applying the well log as a second input to a second machine learning model comprising a convolutional neural network having a plurality of one-dimensional convolution layers that are stacked together, wherein the second machine learning model is configured to predict second markers in the well log based at least in part on a local factor of the well log;

generating a set of predicted well markers by merging at least some of the first markers and at least some of the second markers;

aligning the well log with respect to one or more other well logs based at least in part on the set of predicted well markers;

generating a digital model of a subterranean volume through which the well log extends based at least in part on the well log that was correlated to the other well logs; and

providing the digital model to control drilling parameters as part of a drilling operation based on the digital model, wherein controlling the drilling parameters comprises controlling a drilling trajectory, selecting and controlling a type of equipment utilized in the drilling operation, or controlling parameters of the type of equipment utilized in the drilling operation.

9 . The medium of claim 8 , wherein the operations further comprise:

receiving a training set of one or more well logs having one or more markers associated with the one or more well logs, and

training the first machine learning model and the second machine learning model using the training set.

10 . The medium of claim 9 , wherein the operations further comprise smoothening one or more markers of the training set prior to training the first machine learning model and prior to training the second machine learning model.

11 . The medium of claim 10 , wherein smoothening comprises convolving the one or more markers with a Gaussian kernel.

12 . The medium of claim 8 , wherein the global factor represents a characteristic of the well log as a whole, and wherein the local factor represents a characteristic of an individual layer of a plurality of layers represented in the well log.

13 . A computing system, comprising:

a display device;

one or more processors; and

a memory system including one or more non-transitory computer-readable media storing instructions that, when executed by at least one of the one or more processors, cause the computing system to perform operations, the operations comprising:

receiving a well log generated based upon recordings of measurements taken along an entirety of a well or along a portion of the well via an instrument deployed into the well;

applying the well log as a first input to a first machine learning model comprising an encoder-decoder architecture with skip connections between mirroring layers of an encoder and a decoder, wherein the first machine learning model is configured to predict first markers in the well log based at least in part on a global factor of the well log, wherein the global factor represents a characteristic of the well log as a whole;

applying the well log as a second input to a second machine learning model comprising a convolutional neural network having a plurality of one-dimensional convolution layers that are stacked together, wherein the second machine learning model is configured to predict second markers in the well log based at least in part on a local factor of the well log, wherein the local factor represents a characteristic of an individual layer of a plurality of layers represented in the well log;

generating a set of predicted well markers by merging at least some of the first markers and at least some of the second markers;

aligning the well log with respect to one or more other well logs based at least in part on the set of predicted well markers;

generating a digital model of a subterranean volume through which the well log extends based at least in part on the well log that was correlated to the other well logs; and

displaying the digital model using the display device; and

providing the digital model to control drilling parameters as part of a drilling operation based on the digital model, wherein controlling the drilling parameters comprises controlling a drilling trajectory, selecting and controlling a type of equipment utilized in the drilling operation, or controlling parameters of the type of equipment utilized in the drilling operation.

Assignments (1)

Continuity (2)

References Cited (23)