Drilling operations framework

Abstract

A method can include acquiring data for rig operations that move a drillstring in a borehole in a subsurface geologic region, where the drillstring includes connected stands of drill pipe and a drill bit for drilling into the subsurface geologic region, and where the data include measured depth data, inclination data, mud density data, and measured hook load data; generating an estimated hook load value for a measured depth in the borehole using at least a trained model that receives a portion of the data as associated with the measured depth; performing a comparison between the estimated hook load value and a measured hook load value of the measured hook load data as associated with the measured depth; and, based at least in part on the comparison, determining a level of sticking of the drillstring in the borehole.

What is claimed is: 
     
         1 . A method comprising:
 acquiring data for rig operations that move a drillstring in a borehole in a subsurface geologic region, wherein the drillstring comprises connected stands of drill pipe and a drill bit for drilling into the subsurface geologic region, and wherein the data comprise measured depth data, inclination data, mud density data, and measured hook load data;   generating an estimated hook load value for a measured depth in the borehole using at least a trained model that receives a portion of the data as associated with the measured depth;   performing a comparison between the estimated hook load value and a measured hook load value of the measured hook load data as associated with the measured depth; and   based at least in part on the comparison, determining a level of sticking of the drillstring in the borehole.   
     
     
         2 . The method of  claim 1 , wherein the generating generates the estimated hook load value using a friction factor value. 
     
     
         3 . The method of  claim 2 , comprising determining the friction factor value by comparing a number of estimated hook load values for different friction factors for a span of measured depths to a number of measured hook load values of the measured hook load data for the span of measured depths. 
     
     
         4 . The method of  claim 1 , wherein the trained model receives an inclination value of the inclination data and a mud density value of the mud density data. 
     
     
         5 . The method of  claim 1 , wherein the estimated hook load value depends on a measured depth value, an inclination value, and a mud density value. 
     
     
         6 . The method of  claim 1 , wherein the acquiring acquires real-time data during one or more types of the rig operations. 
     
     
         7 . The method of  claim 6 , wherein the one or more types of the rig operations comprise a pulling out type of rig operation and a running in type of rig operation. 
     
     
         8 . The method of  claim 6 , comprising controlling one or more of the rig operations based at least on the level of sticking. 
     
     
         9 . The method of  claim 1 , wherein the level of sticking comprises a less than micro sticking level and a micro sticking level. 
     
     
         10 . The method of  claim 9 , wherein the micro sticking level is associated with an increased risk of a higher level of sticking. 
     
     
         11 . The method of  claim 1 , comprising generating at least one control instruction associated with the level of sticking. 
     
     
         12 . The method of  claim 11 , wherein the at least one control instruction comprises a control instruction to add an additive to drilling fluid to reduce risk of sticking. 
     
     
         13 . The method of  claim 11 , wherein the at least one control instruction comprises a control instruction to adjust speed of moving the drillstring in the borehole. 
     
     
         14 . The method of  claim 1 , wherein the generating the estimated hook load value comprises using a filter that comprises an input for the measured hook load value and an input for a predicted hook load value. 
     
     
         15 . The method of  claim 14 , wherein the filter comprises a Bayesian type of filter. 
     
     
         16 . The method of  claim 15 , wherein the Bayesian type of filter comprises a Bayesian Kalman filter. 
     
     
         17 . The method of  claim 1 , wherein the generating comprises estimating uncertainty of the estimated hook load value. 
     
     
         18 . The method of  claim 1 , wherein the trained model comprises a Gaussian Process Regression model. 
     
     
         19 . A system comprising:
 a processor;   memory accessible by the processor;   processor-executable instructions stored in the memory and executable to instruct the system to:
 acquire data for rig operations that move a drillstring in a borehole in a subsurface geologic region, wherein the drillstring comprises connected stands of drill pipe and a drill bit for drilling into the subsurface geologic region, and wherein the data comprise measured depth data, inclination data, mud density data, and measured hook load data; 
   generate an estimated hook load value for a measured depth in the borehole using at least a trained model that receives a portion of the data as associated with the measured depth;   perform a comparison between the estimated hook load value and a measured hook load value of the measured hook load data as associated with the measured depth; and   based at least in part on the comparison, determine a level of sticking of the drillstring in the borehole.   
     
     
         20 . One or more computer-readable storage media comprising processor-executable instructions to instruct a computing system to:
 acquire data for rig operations that move a drillstring in a borehole in a subsurface geologic region, wherein the drillstring comprises connected stands of drill pipe and a drill bit for drilling into the subsurface geologic region, and wherein the data comprise measured depth data, inclination data, mud density data, and measured hook load data;   generate an estimated hook load value for a measured depth in the borehole using at least a trained model that receives a portion of the data as associated with the measured depth;   perform a comparison between the estimated hook load value and a measured hook load value of the measured hook load data as associated with the measured depth; and   based at least in part on the comparison, determine a level of sticking of the drillstring in the borehole.