Predictive downlink automation process and artificial intelligence powered system

Abstract

Methods and systems for recommending downlink parameters for successful downlinking are described. In one embodiment, a processor receives one or more data records representing downlinks to a downhole tool, determines whether each respective downlink successfully communicated a command, identifies one or more variables to a machine learning algorithm, trains the machine learning algorithm by identifying correlations between the one or more variables and the downlink success, the machine learning algorithm receives drilling condition data, predicts downlink success probability for each of a plurality of downlink parameter combinations based on drilling condition data, and recommends one of the plurality of downlink parameter combinations.

What is claimed is: 
     
         1 . A method comprising:
 a processor receiving one or more data records representing previous drilling operations, the one or more data records including data representing one or more downlinks to a downhole tool;   the processor determining, for each data record, whether each respective downlink of the one or more downlinks successfully communicated a command to the downhole tool;   the processor identifying one or more variables to a machine learning algorithm, at least one of the one or more variables corresponding to a drilling condition;   the processor training the machine learning algorithm using at least a first subset of the received data and data indicating downlink success, the machine learning algorithm training by identifying correlations between the one or more variables and the downlink success; and   the machine learning algorithm receiving drilling condition data;   the machine learning algorithm predicting downlink success probability for each of a plurality of downlink parameter combinations based on the drilling condition data; and   the machine learning algorithm recommending one of the plurality of downlink parameter combinations for communication with the downhole tool based on the predicted downlink success probability.   
     
     
         2 . The method of  claim 1 , wherein the machine learning algorithm is selected from the group consisting of logical regression, random forest, or gradient boosting. 
     
     
         3 . The method of  claim 1 , wherein the drilling condition data comprises geographic area for drilling, a wellbore diameter, mud rheology, mud density, type of bottom hole assembly, rate of penetration, flow amplitude, rotations per minute amplitude, bit period, shock and vibrations, true vertical depth of a point in a drilling path, measured depth of the point in the drilling path, and stick/slip vibration. 
     
     
         4 . The method of  claim 1 , further comprising, the processor preprocessing the received data. 
     
     
         5 . The method of  claim 4 , wherein preprocessing the data further comprises removing fantom downlinks, encoding categorical values, and scaling numerical features. 
     
     
         6 . The method of  claim 4 , wherein preprocessing the data further comprises appending the received data to include data indicating whether each downlink was successful or unsuccessful. 
     
     
         7 . The method of  claim 1 , wherein training the machine learning algorithm further comprises assigning weights to the one or more variables and tuning hyperparameters using grid search or random search. 
     
     
         8 . The method of  claim 1 , wherein the processor determines, for each data record, whether each respective downlink of the one or more downlinks successfully communicated the command to the downhole tool by reviewing subsequent downlinks to determine whether the subsequent downlinks indicate corrective actions taken due to an unsuccessful downlink. 
     
     
         9 . The method of  claim 1 , further comprising:
 the machine learning algorithm predicting, for each record in a second subset of the received data, a probability of successful downlinking for each of the plurality of downlink parameter combinations based on the drilling condition data included in the second subset of received data.   
     
     
         10 . The method of  claim 1 , further comprising:
 the processor calculating or estimating, for each respective downlink of the one or more downlinks, a rate of penetration, a measured depth, a true vertical depth, and a stick/slip value,   wherein the processor calculates or estimates the rate of penetration by calculating a time duration of a power-up session and dividing a predetermined drilling distance for each power-up session by the time duration of the power-up session,   wherein the processor calculates or estimates the measured depth by multiplying the rate of penetration by the time duration of the powerup session,   wherein the processor calculates or estimates the true vertical depth by multiplying the measured depth for the powerup session and a cosine value of an inclination angle of the downhole tool, and   wherein the processor calculates or estimates the stick/slip value by calculating the difference between a maximum turbine rotations per minute value and a minimum turbine rotations per minute value, dividing the difference by a mean turbine rotations per minute value to generate a quotient, and multiplying the quotient by 100 to generate an estimated stick/slip value.   
     
     
         11 . The method of  claim 1 , wherein drilling condition data is generated by one or more sensors included in the downhole tool during a drilling operation. 
     
     
         12 . The method of  claim 1 , wherein drilling condition data comprises a planned well drill path, the planned well drill path including a plurality of points having a measured depth coordinate and a true vertical depth coordinate. 
     
     
         13 . A system comprising:
 a storage device configured to store a machine learning algorithm; and   a processor in communication with the storage device and configured to:   receive one or more data records representing previous drilling operations, the one or more data records including data representing one or more downlinks to a downhole tool;   determine, for each data record, whether each respective downlink of the one or more downlinks successfully communicated a command to the downhole tool;   identify one or more variables to a machine learning algorithm, at least one of the one or more variables corresponding to a drilling condition;   train the machine learning algorithm using at least a first subset of the received data and data indicating downlink success, the machine learning algorithm training by identifying correlations between the one or more variables and the downlink success; and   execute the machine learning algorithm, where the machine learning algorithm is configured to:
 receive drilling condition data; 
 predict downlink success probability for each of a plurality of downlink parameter combinations based on the drilling condition data; and 
 recommend one of the plurality of downlink parameter combinations for communication with the downhole tool based on the predicted downlink success probability. 
   
     
     
         14 . The system of  claim 13 , wherein the machine learning algorithm is selected from the group consisting of logical regression, random forest, or gradient boosting. 
     
     
         15 . The system of  claim 13 , wherein the drilling condition data comprises geographic area for drilling, a wellbore diameter, mud rheology, mud density, type of bottom hole assembly, rate of penetration, flow amplitude, rotations per minute amplitude, bit period, shock and vibrations, true vertical depth of a point in a drilling path, measured depth of the point in the drilling path, and stick/slip vibration. 
     
     
         16 . The system of  claim 13 , wherein the processor is further configured to preprocess the received data. 
     
     
         17 . The system of  claim 16 , wherein the processor is further configured to remove fantom downlinks, encode categorical values, and scale numerical features as part of a preprocessing function. 
     
     
         18 . The system of  claim 16 , wherein the processor is further configured to append the received data to include data indicating whether each downlink was successful or unsuccessful as part of a preprocessing function. 
     
     
         19 . The system of  claim 13 , wherein the processor is further configured to assign weights to the one or more variables and tuning hyperparameters using grid search or random search as part of training the machine learning algorithm. 
     
     
         20 . The system of  claim 13 , wherein the processor is further configured to review subsequent downlinks to determine whether the subsequent downlinks indicate corrective actions taken due to an unsuccessful downlink in order to determine whether each respective downlink of the one or more downlinks successfully communicated the command to the downhole tool. 
     
     
         21 . The system of  claim 13 , wherein the machine learning algorithm is further configured to:
 predict, for each record in a second subset of the received data, a probability of successful downlinking for each of the plurality of downlink parameter combinations based on the drilling condition data included in the second subset of received data.   
     
     
         22 . The system of  claim 13 , wherein the processor is further configured to:
 calculate or estimate, for each respective downlink of the one or more downlinks, a rate of penetration, a measured depth, a true vertical depth, and a stick/slip value,   wherein the processor calculates or estimates the rate of penetration by calculating a time duration of a power-up session and dividing a predetermined drilling distance for each power-up session by the time duration of the power-up session,   wherein the processor calculates or estimates the measured depth by multiplying the rate of penetration by the time duration of the powerup session,   wherein the processor calculates or estimates the true vertical depth by multiplying the measured depth for the powerup session and a cosine value of an inclination angle of the downhole tool, and   wherein the processor calculates or estimates the stick/slip value by calculating the difference between a maximum turbine rotations per minute value and a minimum turbine rotations per minute value, dividing the difference by a mean turbine rotations per minute value to generate a quotient, and multiplying the quotient by 100 to generate an estimated stick/slip value.   
     
     
         23 . The system of  claim 13 , wherein drilling condition data is generated by one or more sensors included in the downhole tool during a drilling operation. 
     
     
         24 . The system of  claim 13 , wherein drilling condition data comprises a planned well drill path, the planned well drill path including a plurality of points having a measured depth coordinate and a true vertical depth coordinate. 
     
     
         25 . A non-transitory machine-readable medium comprising instructions, which, when executed by one or more processors, cause the one or more processors to perform the following operations:
 receive one or more data records representing previous drilling operations, the one or more data records including data representing one or more downlinks to a downhole tool;   determine, for each data record, whether each respective downlink of the one or more downlinks successfully communicated a command to the downhole tool;   identify one or more variables to a machine learning algorithm, at least one of the one or more variables corresponding to a drilling condition;   train the machine learning algorithm using at least a first subset of the received data and data indicating downlink success, the machine learning algorithm trained by identifying correlations between the one or more variables and the downlink success; and receive, by the machine learning algorithm, drilling condition data;   predict, by the machine learning algorithm, downlink success probability for each of a plurality of downlink parameter combinations based on the drilling condition data; and   recommend, by the machine learning algorithm, one of the plurality of downlink parameter combinations for communication with the downhole tool based on the predicted downlink success probability.