Spectral analysis and machine learning for determining cluster efficiency during fracking operations
This disclosure presents systems, methods, and apparatus for determining cluster efficiency during hydraulic fracturing, the method comprising: measuring acoustic vibrations in fracking fluid in a fracking wellhead, circulating fluid line, or standpipe of a well; converting the acoustic vibrations into an electrical signal in a time domain; recording the electrical signal to memory; analyzing the electrical signal in the time domain for a window of time and identifying two amplitude peaks corresponding to a fracture initiation; measuring a time between the two amplitude peaks; dividing the time by two to give a result; multiplying the result by a speed of sound in the fracking fluid to give a distance between the fracture initiation and a plug at an end of a current fracking stage of the well; and returning a location of the fracture initiation to an operator based on the distance between the fracture initiation and the plug.
1. A non-transitory, tangible computer readable storage medium, encoded with processor readable instructions to perform a method for determining cluster efficiency during hydraulic fracturing operations, the method comprising:
converting the acoustic vibrations into an electrical signal in a time domain;
recording the electrical signal to a memory;
identifying a fracture initiation from the electrical signal in the time domain via identification of two amplitude peaks occurring within a threshold period of time of each other;
measuring a time between the two amplitude peaks;
dividing the time by two to give a result;
multiplying the result by a speed of sound in the fracking fluid to give a distance between the fracture initiation and a plug at an end of a current fracking stage of the well; and
using the distance to create a cluster map of fracture initiations in the stage of the well.
2. The non-transitory, tangible computer readable storage medium of claim 1 , wherein the measuring is performed by a sensor coupled to the wellhead, circulating fluid line, or standpipe.
3. The non-transitory, tangible computer readable storage medium of claim 2 , wherein the sensor samples at greater than 1 kHz.
4. The non-transitory, tangible computer readable storage medium of claim 2 , wherein the sensor is an acoustic sensor.
5. The non-transitory, tangible computer readable storage medium of claim 1 , wherein the two amplitude peaks are identified based on exceeding an amplitude threshold.
6. The non-transitory, tangible computer readable storage medium of claim 1 , wherein the two amplitude peaks are identified based on a combination of amplitude and width.
7. The non-transitory, tangible computer readable storage medium of claim 1 , wherein the measuring is performed by a sensor, and wherein the sensor is configured to be in contact with the fracking fluid in the well or with a surface of the circulating fluid line or the standpipe at the wellhead.
8. The non-transitory, tangible computer readable storage medium of claim 1 , wherein the method further comprises: analyzing the electrical signal for the window of time in the frequency domain to identify the two amplitude peaks.
9. The non-transitory, tangible computer readable storage medium of claim 1 , wherein the method further comprises: performing the identifying based also on analysis of pressure sensor data during the window of time.
10. The non-transitory, tangible computer readable storage medium of claim 1 , wherein the analyzing is performed on a locally-hosted machine-learning system.
11. The non-transitory, tangible computer readable storage medium of claim 1 , wherein the electrical signal in the time domain is transported via a large area network to a remote server hosting a machine-learning system configured to perform the analyzing.
12. The non-transitory, tangible computer readable storage medium of claim 1 , wherein the method further comprises: adjusting parameters of subsequent fracking operations to improve cluster efficiency of fracture initiations.
13. The non-transitory, tangible computer readable storage medium of claim 1 , wherein the method further comprises: storing the distance from the fracture initiation to the plug with other distances from prior fracture initiations to the plug, and referenced to different plug positions such that a mapping of fracture initiations in clusters associated with each plug position can be determined for a fracking stage.
14. The non-transitory, tangible computer readable storage medium of claim 1 , wherein measuring a time between the two amplitude peaks comprises measuring a time between a maximum of a first of the two peaks and a maximum of a second of the two peaks.
15. The non-transitory, tangible computer readable storage medium of claim 1 , wherein the speed of sound in the fracking fluid is a function of arrival time of a primary wave from the fracture initiation and arrival time of a reflected wave from the fracture initiation after reflecting off the plug.
16. The non-transitory, tangible computer readable storage medium of claim 15 , wherein the speed of sound, S, in the fracking fluid is:
S =( G+P )+2( F+E )/( A 2 −A 1 ),
where G is a firing depth of a perforation gun, P is a depth of the plug, F is a distance between the sensor and the wellhead, E is a distance from the sensor to a fluid end at the surface of the well, A 1 is the arrival time of the primary wave from the fracture initiation, and A 2 is the arrival time of the reflected wave from the fracture initiation and reflected off the plug.
17. The non-transitory, tangible computer readable storage medium of claim 1 , wherein the electrical signal is transformed from the time domain to the frequency domain for a window of time, and the two amplitude peaks are identified by comparing the electrical signal in the frequency domain for the window of time to frequency domain data associated with previous pairs of amplitude peaks corresponding to previous fracture initiations.
18. A non-transitory, tangible computer readable storage medium, encoded with processor readable instructions to perform a method of more evenly dispersing fractures within fracture clusters during hydraulic fracturing operations, the method comprising:
pumping fracking fluid into a stage of a well;
measuring acoustic vibrations in fracking fluid in a wellhead, circulating fluid line, or standpipe of the well;
converting the acoustic vibrations into an electrical signal in a time domain;
recording the electrical signal to a memory;
identifying a fracture initiation from a current frequency domain spectrum via a machine-learning system trained on previous frequency domain spectra measured as a result of previous fracture initiations and previously classified by the machine-learning system;
analyzing the electrical signal in the time domain during a window of time and identifying two amplitude peaks corresponding to the fracture initiation found in the current frequency domain spectrum;
measuring a time between the two amplitude peaks;
dividing the time by two to give a result;
multiplying the result by a speed of sound in fracking fluid thought to be near the fracture initiation to give a distance from the fracture initiation to a plug; and
using the distance to create a cluster map of fracture initiations in the stage of the well; and
adjusting a parameter of the hydraulic fracturing operations based on the cluster map to achieve more even dispersion of fractures in a subsequent stage.
19. The non-transitory, tangible computer readable storage medium of claim 18 , wherein the measuring is performed by a sensor coupled to the wellhead, the circulating fluid line, or the standpipe.
20. The non-transitory, tangible computer readable storage medium of claim 19 , wherein the sensor samples at greater than 1 kHz.
21. The non-transitory, tangible computer readable storage medium of claim 19 , wherein the sensor is an acoustic sensor.
22. The non-transitory, tangible computer readable storage medium of claim 18 , wherein the two amplitude peaks are identified based on exceeding an amplitude threshold.
23. The non-transitory, tangible computer readable storage medium of claim 18 , wherein the two amplitude peaks are identified based on a combination of amplitude and width.
24. The non-transitory, tangible computer readable storage medium of claim 18 , wherein the two amplitude peaks are identified based on a frequency of the two amplitude peaks.
25. The non-transitory, tangible computer readable storage medium of claim 18 , wherein the measuring is performed by a sensor, and wherein the sensor is configured to be in contact with the fracking fluid in the well or with a surface of the circulating fluid line or a surface of the standpipe at the wellhead.
26. The non-transitory, tangible computer readable storage medium of claim 18 , wherein the method further comprises: performing the identifying based also on analysis of pressure sensor data during the window of time.
27. The non-transitory, tangible computer readable storage medium of claim 18 , wherein the analyzing is performed on a locally-hosted machine-learning system.
28. The non-transitory, tangible computer readable storage medium of claim 18 , wherein the electrical signal in the time domain is transported via a large area network to a remote server hosting a machine-learning system configured to perform the analyzing.