IP Library Granted Patent US 12680440
Granted Patent B2
US 12680440 · App. 19/062,535 · Granted Jul 14, 2026

Drilling operations framework

Inventors: Sreelekshmi Jayalekshmi (Cambridge, GB); Sai Venkatakrishnan Sankaranarayanan (Cambridge, GB)
Assignee: Schlumberger Technology Corporation
E21B44/00E21B2200/20E21B2200/22
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 12680440
App. No.
19/062,535
Granted
Jul 14, 2026
Kind
B2
Abstract

A method can include acquiring real-time data during rig operations that include rig operations for drilling a borehole in a subsurface geologic region using a drillstring that includes a drill bit, where the drillstring includes connected stands of drill pipe, and where the real-time data include a measured post-connection hook load value; predicting a post-connection hook load value using a trained model; estimating a post-connection hook load value using a filter that includes an input for the measured post-connection hook load value and an input for the predicted post-connection hook load value; and controlling drilling of the borehole based at least in part on the estimated post-connection hook load value.

Claims (36)

1 . A method comprising:

acquiring real-time data during rig operations that comprise rig operations for drilling a borehole in a subsurface geologic region using a drillstring that comprises a drill bit, wherein the drillstring comprises connected stands of drill pipe, and wherein the real-time data comprises a measured post-connection hook load value;

predicting a post-connection hook load value using a trained model;

estimating a post-connection hook load value using a Kalman filter that comprises i) an input for the measured post-connection hook load value and ii) an input for the predicted post-connection hook load value and iii) an output representing the estimated post-connection hook load; and

controlling drilling of the borehole based at least in part on the estimated post-connection hook load value, wherein the controlling drilling comprises lowering the drillstring to contact the drill bit with a bottom of the borehole while the drill bit is rotating,

wherein the predicted post-connection hook load value comprises model uncertainty, and wherein uncertainty in the estimated post-connection hook load output from the Kalman filter is less than the model uncertainty.

2 . The method of claim 1 , wherein the estimated post-connection hook load value is utilized as a tare value.

3 . The method of claim 1 , wherein the measured post-connection hook load value comprises observed uncertainty that includes temporal uncertainty as to a rig state, and wherein the uncertainty in the estimated post-connection hook load output from the filter is less than the observed uncertainty.

4 . The method of claim 3 , wherein the rig state comprises an off bottom, free rotating state of the drillstring.

5 . The method of claim 4 , wherein the rig state further comprises a circulation state wherein drilling fluid is circulating in the borehole.

6 . The method of claim 3 , wherein the observed uncertainty further includes sensor uncertainty due to sensor-based measurement noise.

7 . The method of claim 1 , wherein the trained model comprises a Gaussian Process model.

8 . The method of claim 7 , wherein the Gaussian Process model comprises a Gaussian Process Regression model.

9 . The method of claim 1 , wherein the trained model comprises inputs for measured depth, inclination and mud density as well as an output for the predicted post-connection hook load, wherein the output represents an off bottom rotating weight for the drillstring with drilling fluid circulating in the borehole.

10 . The method of claim 1 , comprising training a model to generate the trained model wherein the training comprises accessing data for offset wells offset from a rig site of the rig operations.

11 . The method of claim 1 , wherein the Kalman filter includes a prediction process and an update process for output of a state estimate as well as feedback for the prediction process in a subsequent iteration, wherein the prediction process is configured to predict a post-connection hook load value along with uncertainty.

12 . The method of claim 1 , wherein the Kalman filter comprises a Bayesian Kalman filter.

13 . The method of claim 1 , comprising generating a graphical user interface that comprises at least one indicator of the uncertainty in the estimated post-connection hook load as a function of depth in the borehole.

14 . The method of claim 1 , wherein the measured post-connection hook load value corresponds to an operation that adds a stand of drill pipe to the drillstring and the output of the filter represents the estimated post-connection hook load for the stand of drill pipe added to the drillstring.

15 . The method of claim 1 , wherein the controlling drilling further comprises controlling weight on the drill bit based at least in part on the estimated post-connection hook load value.

16 . The method of claim 1 , wherein the rig operations use slips assembly configured to selectively engage and release the drillstring, and an operation corresponding to the measured post-connection hook load value involves the drillstring being released by the slips assembly and a downhole tool being lowered to be at a bottom hole location in the borehole.

17 . A system comprising:

a processor;

memory accessible by the processor; and

processor-executable instructions stored in the memory and executable to instruct the system to:

acquire real-time data during rig operations that comprise rig operations for drilling a borehole in a subsurface geologic region using a drillstring that comprises a drill bit, wherein the drillstring comprises connected stands of drill pipe, and wherein the real-time data comprises a measured post-connection hook load value;

predict a post-connection hook load value using a trained model;

estimate a post-connection hook load value using a Kalman filter that comprises i) an input for the measured post-connection hook load value and ii) an input for the predicted post-connection hook load value and iii) an output representing the estimated post-connection hook load; and

control drilling of the borehole based at least in part on the estimated post-connection hook load value, wherein the control involves lowering the drillstring to contact the drill bit with a bottom of the borehole while the drill bit is rotating,

wherein the predicted post-connection hook load value comprises model uncertainty, and wherein uncertainty in the estimated post-connection hook load output from the Kalman filter is less than the model uncertainty.

18 . One or more computer-readable storage media comprising processor-executable instructions to instruct a computing system to:

acquire real-time data during rig operations that comprise rig operations for drilling a borehole in a subsurface geologic region using a drillstring that comprises a drill bit, wherein the drillstring comprises connected stands of drill pipe, and wherein the real-time data comprises a measured post-connection hook load value;

predict a post-connection hook load value using a trained model;

estimate a post-connection hook load value using a Kalman filter that comprises i) an input for the measured post-connection hook load value and ii) an input for the predicted post-connection hook load value and iii) an output representing the estimated post-connection hook load; and

control drilling of the borehole based at least in part on the estimated post-connection hook load value, wherein the control involves lowering the drillstring to contact the drill bit with a bottom of the borehole while the drill bit is rotating,

wherein the predicted post-connection hook load value comprises model uncertainty, and wherein uncertainty in the estimated post-connection hook load output from the Kalman filter is less than the model uncertainty.