Model generation for real-time rate of penetration prediction

Patent metadata

Field	Value
Publication number	US-10657441-B2
Application number	US-201515551204-A
Country	US
Kind code	B2
Filing date	Apr 1, 2015
Priority date	Apr 1, 2015
Publication date	May 19, 2020
Grant date	May 19, 2020

Abstract

An example method includes receiving raw data sets containing drilling parameter and operating condition values generated during subterranean drilling operations. The raw data sets may be separated into training data sets based, at least in part, on the types of the subterranean drilling operations. At least one predictive model may be generated based, at least in part, on at least one training data set. The at least one predictive model may determine a rate of penetration (ROP) for a drilling operation of the same type to which the at least one training data set corresponds.

First claim

What is claimed is:

1. A method, comprising:
receiving raw data sets comprising dynamic data and static data, wherein the dynamic data comprises drilling parameter and operating condition values generated during subterranean drilling operations, wherein the static data is indicative of one or more types of the subterranean drilling operations that generated the dynamic data;
separating the raw data sets into training data sets based, at least in part, on the one or more types of the subterranean drilling operations identified in the static data of the raw data sets;
generating at least one predictive model based, at least in part, on at least one training data set of the training data sets, wherein the at least one predictive model determines a rate of penetration (ROP) for the one or more types to which the at least one training data set corresponds, wherein generating the at least one predictive model comprises for each training data set of the training data sets generating a different context-specific predictive model associated with the static data used to generate the at least one predictive model.

2. The method of claim 1 , further comprising the step of reducing dimensionality of at least one of the training data sets using at least one feature extraction technique before the at least one of the training data sets is used for the at least one predictive model.

3. The method of claim 2 , wherein the at least one feature extraction technique comprises at least one of a principal component analysis, a partial least squares regression, an independent component analysis, an isomap, and an autoencoder.

4. The method of claim 1 , wherein generating at least one predictive model based, at least in part, on at least one training data set comprises training a learning algorithm using the at least one training data set.

5. The method of claim 4 , wherein the learning algorithm comprises at least one of a decision tree, a Bayesian belief network, a genetic algorithm, an artificial neural network, and a support vector machines.

6. The method of claim 5 , wherein training a learning algorithm using the at least one training data set comprises determining at least one parameter of the learning algorithm using at least one of a grid search, a randomized parameter optimization, and a linear search.

7. The method of claim 1 , further comprising reducing a number of data entries in at least one of the training data sets based, at least in part, on a pre-determined threshold.

8. The method of claim 1 , further comprising at least one of thresholding, ROP filtering, averaging, and normalizing the raw data.

9. The method of claim 1 , further comprising separating the raw data sets into the dynamic data and the static data.

10. The method of claim 9 , wherein the one or more types to which the at least one training data set corresponds is based, at least in part, on the static data.

11. The method of claim 1 , further comprising removing at least some data entries of the raw data sets based, at least in part, on ROP values within the data entries.

12. The method of claim 1 , further comprising determining a ROP for a drilling operation using the model and altering at least one drilling parameter of the drilling operation based, at least in part, on the determined ROP.

13. The method of claim 1 ,
wherein receiving raw data sets containing drilling parameter and operating condition values generated during subterranean drilling operations comprises receiving raw data sets containing numerical values corresponding to at least one of a weight on bit (WOB), rotary speed, drill bit rotations per minute (RPM), hook load, surface torque, torque on bit, downhole mud flow rate, return mud flow rate, stand pipe pressure (SPP), and ROP; and
wherein separating the raw data sets into training data sets based, at least in part, on the one or more types of the subterranean drilling operations comprises separating the raw data sets based, at least in part, on a formation lithology, a drill bit type, a drill bit size, a drilling assembly type, and a well inclination of the subterranean drilling operations.

14. A non-transitory computer readable medium containing a set of instructions that, when executed by a processor of an information handling system, cause the processor to
receive raw data sets comprising dynamic data and static data, wherein the dynamic data comprises drilling parameter and operating condition values generated during subterranean drilling operations, wherein the static data is indicative of one or more types of the subterranean drilling operations that generated the dynamic data;
separate the raw data sets into training data sets based, at least in part, on the one or more types of the subterranean drilling operations identified in the static data of the raw data sets;
generate at least one predictive model based, at least in part, on at least one training data set of the training data sets, wherein the at least one predictive model determines a rate of penetration (ROP) for the one or more types to which the at least one training data set corresponds, wherein generating the at least one predictive model comprises for each training data set of the training data sets generating a different context-specific predictive model associated with the static data used to generate the at least one predictive model.

15. The non-transitory computer readable medium of claim 14 , wherein the set of instructions further cause the processor to reduce dimensionality of at least one of the training data sets using at least one feature extraction technique before the at least one of the training data sets is used for the at least one predictive model.

16. The non-transitory computer readable medium of claim 15 , wherein the at least one feature extraction technique comprises at least one of a principal component analysis, a partial least squares regression, an independent component analysis, an isomap, and an autoencoder.

17. The non-transitory computer readable medium of claim 14 , wherein the set of instructions that cause the processor to generate at least one predictive model based, at least in part, on at least one training data set further cause the processor to train a learning algorithm using the at least one training data set.

18. The non-transitory computer readable medium of claim 17 , wherein the learning algorithm comprises at least one of a decision tree, a Bayesian belief network, a genetic algorithm, an artificial neural network, and a support vector machines.

19. The non-transitory computer readable medium of claim 18 , wherein the set of instructions that cause the processor to train a learning algorithm using the at least one training data set further cause the processor to determine at least one parameter of the learning algorithm using at least one of a grid search, a randomized parameter optimization, and a linear search.

20. The non-transitory computer readable medium of claim 14 , wherein the set of instructions further cause the processor to reduce a number of data entries in at least one of the training data sets based, at least in part, on a pre-determined threshold.

21. The non-transitory computer readable medium of claim 14 , wherein the set of instructions further cause the processor to threshold, ROP filter, average, and normalize the raw data.

22. The non-transitory computer readable medium of claim 14 , wherein the set of instructions further cause the processor to separate the raw data

Assignees

• Landmark Graphics Corp

Inventors

• Dursun Serkan
• Tran Nhon Ai T
• Toti Giulia

Classifications

• G06F9/455

  Emulation; Interpretation; Software simulation, e.g. virtualisation or emulation of application or operating system execution engines · CPC title

• E21B44/00 Primary

  Automatic control systems specially adapted for drilling operations, i.e. self-operating systems which function to carry out or modify a drilling operation without intervention of a human operator, e.g. computer-controlled drilling systems; Systems specially adapted for monitoring a plurality of drilling variables or conditions · CPC title

• G06G7/48

  Analogue computers for specific processes, systems or devices, e.g. simulators · CPC title

• G06N3/08 Primary

  Learning methods · CPC title

• E21B45/00

  Measuring the drilling time or rate of penetration · CPC title