Predictive pump maintenance based upon utilization and operating conditions

Patent metadata

Field	Value
Publication number	US-12560078-B2
Application number	US-202418610893-A
Country	US
Kind code	B2
Filing date	Mar 20, 2024
Priority date	Nov 17, 2021
Publication date	Feb 24, 2026
Grant date	Feb 24, 2026

Abstract

A computer implemented method of predicting a future maintenance event of a pumping equipment on a wellbore pumping unit comprising loading a pump usage log and a pump maintenance log into a predictive maintenance model. The predictive maintenance model is trained by a machine learning process with a historical database of completed pumping jobs. The predictive maintenance model determines a probability of a future maintenance event in response to the current pump usage. The unit controller displays an alert of the remaining pump life in comparison to a threshold value for a recommended pump maintenance period or a required pump maintenance period.

First claim

What is claimed is:

1 . A wellbore servicing method comprising:
transporting a pump unit to a wellsite, the pump unit comprising pumping equipment and a unit controller comprising a processor, a non-transitory memory, and an input-output device: fluidically connecting the pump unit to a wellbore; operating the pumping equipment so as to perform a pumping operation to deliver a fluid into the wellbore; collecting one or more datasets associated with the pumping operation; retrieving one or more historical datasets associated with the pumping equipment; determining a fitness indicator for the pumping equipment based upon the one or more datasets associated with the pumping operation and the one or more historical datasets, wherein the fitness indicator comprises a probability of an imminent maintenance event of the pumping equipment; and outputting, via the unit controller, indicia of the fitness indicator for the pumping equipment via the input-output device.

2 . The method of claim 1 , wherein the one or more historical datasets comprise historical usage data, historical maintenance data, or combinations thereof.

3 . The method of claim 1 , wherein the one or more datasets associated with the pumping operation comprise data indicative of a flowrate of the fluid, data indicative of a pressure of the fluid, data indicative of a volume of the fluid, data indicative of a parameter of the fluid, or combinations thereof.

4 . The method of claim 1 , wherein the fitness indicator further comprises a pump life value indicative of predicted usage of the pumping equipment before the imminent maintenance event.

5 . The method of claim 1 , wherein the fitness indicator for the pumping equipment is determined via a fitness determination model.

6 . The method of claim 5 , wherein the fitness determination model is trained by:
retrieving, by a machine learning process, a plurality of historical pump usage records from one or more pumping operations respectively associated with a plurality of pumping equipment; retrieving, by the machine learning process, a plurality of historical indicia of fitness respectively associated with the plurality of pumping equipment; and training, by the machine learning process, the fitness determination model.

7 . The method of claim 1 , further comprising adjusting, via the unit controller, the operation of the pumping equipment during the pumping operation.

8 . The method of claim 1 , further comprising adjusting, via the unit controller, the operation of the pumping equipment subsequent to the pumping operation.

9 . The method of claim 1 , wherein the indicia of the fitness indicator for the pumping equipment comprises a visual cue, and audible cue, or both.

10 . A system of wellbore pumping unit, comprising:
a wellbore pumping unit comprising a mixing system comprising a supply pump, a main pump, a plurality of sensors, and an input-output device; a unit controller comprising a processor, a non-transitory memory, and an input-output, configured to:
operate a pumping equipment so as to perform a pumping operation to deliver a fluid into a wellbore, wherein the unit controller comprises a processor, a non-transitory memory, and an input-output device;
collect one or more datasets associated with the pumping operation;
retrieve one or more historical datasets associated with the pumping equipment;
determine a fitness indicator for the pumping equipment based upon the one or more datasets associated with the pumping operation and the one or more historical datasets, wherein the fitness indicator comprises a probability of an imminent maintenance event of the pumping equipment; and
output indicia of the fitness indicator for the pumping equipment via the input-output device.

11 . The system of claim 10 , wherein the one or more historical datasets comprise historical usage data, historical maintenance data, or combinations thereof.

12 . The system of claim 10 , wherein the one or more datasets associated with the pumping operation comprise data indicative of a flowrate of the fluid, data indicative of a pressure of the fluid, data indicative of a volume of the fluid, data indicative of a parameter of the fluid, or combinations thereof.

13 . The system of claim 10 , wherein the fitness indicator further comprises a pump life value indicative of predicted usage of the pumping equipment before the imminent maintenance event.

14 . The system of claim 10 , wherein the fitness indicator for the pumping equipment is determined via a fitness determination model.

15 . The system of claim 14 , wherein the fitness determination model is trained by:
retrieving, by a machine learning process, a plurality of historical pump usage records from one or more pumping operations respectively associated with a plurality of pumping equipment; retrieving, by the machine learning process, a plurality of historical indicia of fitness respectively associated with the plurality of pumping equipment; and training, by the machine learning process, the fitness determination model.

16 . The system of claim 10 , further comprising adjusting, via the unit controller, the operation of the pumping equipment during the pumping operation.

17 . The system of claim 10 , further comprising adjusting, via the unit controller, the operation of the pumping equipment subsequent to the pumping operation.

18 . The system of claim 10 , wherein the indicia of the fitness indicator for the pumping equipment comprises a visual cue, and audible cue, or both.

Assignees

• Halliburton Energy Services Inc

Inventors

• Neal Iii Charles E
• Williams Derek R
• Sneed Craig Allen
• Oliver Sunny L
• Blaschke Keith Edward

Classifications

• G06F18/214

  Generating training patterns; Bootstrap methods, e.g. bagging or boosting · CPC title

• G06N20/00

  Machine learning · CPC title

• E21B2200/20

  Computer models or simulations, e.g. for reservoirs under production, drill bits · CPC title

• F04B51/00

  Testing machines, pumps, or pumping installations · CPC title

• F04B49/065

  and making use of computers · CPC title