Machine learning mud pulse recognition networks

Patent metadata

Field	Value
Publication number	US-11725505-B2
Application number	US-202217665999-A
Country	US
Kind code	B2
Filing date	Feb 7, 2022
Priority date	Dec 22, 2020
Publication date	Aug 15, 2023
Grant date	Aug 15, 2023

Abstract

This disclosure presents a process for communications in a borehole containing a fluid or drilling mud, where a conventional mud pulser can be utilized to transmit data to a transducer. The transducer, or a communicatively coupled computing system, can perform pre-processing steps to correct the received data using an average of a moving time window of the received data, and then normalize the corrected data. The corrected data can then be utilized as inputs into a machine learning mud pulse recognition network where the data can be classified and an ideal or clean pulse waveform can be overlaid the corrected data. The overlay and the corrected data can be fed into a conventional decoder or decoded by the disclosed process. The decoded data can then be communicated to another system and used as inputs, such as to a well site controller to enable adjustments to well site operation parameters.

First claim

What is claimed is:

1. A method, comprising:
obtaining a mud pulse data transmission from a mud pulser located in a borehole;
processing the mud pulse data transmission to generate normalized corrected data; and
generating SNR corrected data from the mud pulse data transmission utilizing a machine learning mud pulse recognition network (MPRN) and the normalized corrected data as an input to the machine learning MPRN.

2. The method as recited in claim 1 , further comprising generating a data transmission overlay mapping utilizing the machine learning MPRN and the normalized corrected data as the input.

3. The method as recited in claim 2 , wherein generating the SNR corrected data includes applying the data transmission overlay mapping to the normalized corrected data.

4. The method as recited in claim 1 , wherein the processing includes averaging the mud pulse data transmission using a moving time window, determining a corrected data utilizing the averaging and the mud pulse data transmission, and generating the normalized corrected data by normalizing the corrected data.

5. The method as recited in claim 4 , wherein a length of the moving time window is determined by a received input parameter that is modified by one or more environmental parameters of the borehole.

6. The method as recited in claim 4 , wherein the determining the corrected data includes subtracting a running average from the averaging of the mud pulse data transmission.

7. The method as recited in claim 1 , wherein the obtaining includes receiving, from the mud pulser located in the borehole, the mud pulse data transmission at a transducer.

8. The method as recited in claim 1 , further comprising transmitting the SNR corrected data to a decoder.

9. The method as recited in claim 8 , further comprising:
cleaning the SNR corrected data utilizing a classification of an output of the machine learning MPRN to generate a clean corrected data; and
decoding the clean corrected data utilizing the decoder.

10. The method as recited in claim 8 , further comprising decoding the SNR corrected data at the decoder and using the decoded SNR corrected data in a well system operation.

11. A computer program product having a series of operating instructions stored on a non-transitory computer-readable medium that directs a data processing apparatus when executed thereby to perform operations to decode mud pulse data transmissions, the operations comprising:
obtaining a mud pulse data transmission from a mud pulser located in a borehole;
processing the mud pulse data transmission to generate normalized corrected data; and
generating SNR corrected data from the mud pulse data transmission utilizing a machine learning mud pulse recognition network (MPRN) and the normalized corrected data as an input to the machine learning MPRN.

12. The computer program product as recited in claim 11 , the operations further comprising selecting one or more preambles from the normalized corrected data and training the machine learning MPRN utilizing the one or more preambles.

13. The computer program product as recited in claim 12 , wherein the training uses big data machine learning utilizing recorded data retrieved at an end of a drilling job or operation segment.

14. The computer program product as recited in claim 13 , wherein the big data machine learning correlates neural network training with environment parameters of the borehole.

15. The computer program product as recited in claim 12 , wherein the recorded data is communicated to the mud pulser at the end of the drilling job or the operation segment.

16. The computer program product as recited in claim 11 , the operations further comprising generating a data transmission overlay mapping utilizing the machine learning MPRN and the normalized corrected data as the input.

17. The computer program product as recited in claim 16 , wherein generating the SNR corrected data includes applying the data transmission overlay mapping to the normalized corrected data.

18. The computer program product as recited in claim 11 , wherein the processing includes averaging the mud pulse data transmission using a moving time window, determining a corrected data utilizing the averaging and the mud pulse data transmission, and generating the normalized corrected data by normalizing the corrected data.

19. The computer program product as recited in claim 18 , wherein a length of the moving time window is determined by a received input parameter that is modified by one or more pulse parameters of the mud pulse data transmission.

20. The computer program product as recited in claim 18 , wherein the determining the corrected data includes subtracting a running average from the averaging of the mud pulse data transmission.

Assignees

• Halliburton Energy Services Inc

Inventors

• Chen Dingding
• Gao Li
• Lie Joni Polili
• Badrinarayanan Paravastu
• Shah Faisal Farooq
• Pillai Bipin K
• Thottempudi Murali Krishna

Classifications

• G06N3/0499

  Feedforward networks · CPC title

• G06N3/09

  Supervised learning · CPC title

• E21B47/18 Primary

  through the well fluid {, e.g. mud pressure pulse telemetry} · CPC title

• E21B2200/22

  Fuzzy logic, artificial intelligence, neural networks or the like · CPC title

• G06N20/00

  Machine learning · CPC title