Pump integrity detection, monitoring and alarm generation
An embodiment includes a method of monitoring a fluid pump that includes receiving time domain measurement data indicating vibrations occurring in a fluid pump, and filtering the measurement data to remove measurement data components having frequencies below a threshold frequency, the removed measurement data components associated with cyclical motions of the fluid pump. The method also includes dividing the filtered measurement data into a plurality of subsets, each subset corresponding to a pump cycle, and estimating a peak count for each subset, the peak count being a number of peaks having an amplitude that exceeds a selected amplitude threshold, the amplitude threshold associated with impacts between internal components of the pump. The method further includes comparing the peak count with an expected peak count, and determining whether the pump is in a condition selected from at least one of a wear condition and a failure condition based on the comparison.
1. A method of monitoring a fluid pump, the method comprising:
receiving time domain measurement data from an accelerometer disposed on a valve of a fluid pump, the measurement data comprising vibrations occurring in the fluid pump created by impacts of the valve;
filtering the measurement data to remove measurement data components that are associated with cyclical motions of the fluid pump and that have frequencies below a threshold frequency;
dividing the filtered measurement data into a plurality of subsets, each subset corresponding to a pump cycle;
estimating a peak count for each subset, the peak count being a number of peaks having an amplitude that exceeds a selected amplitude threshold, the amplitude threshold associated with the impacts of the valve;
comparing the peak count with an expected peak count;
determining whether the pump is in a condition selected from at least one of a wear condition and a failure condition based on the comparison;
indicating a wear condition when it is determined that the pump is in a wear condition, and indicating a failure condition when it is determined that the pump is in a failure condition; and
operating the pump based on what is indicated.
2. The method of claim 1 , wherein the time domain measurement data is real time data collected during a hydraulic stimulation operation, and the determining is performed in real time during the operation.
3. The method of claim 1 , further comprising using an inductive proximity sensor to identify a speed and temporal length of a cycle of the pump.
4. The method of claim 1 , wherein the step of filtering isolates signals at frequencies associated with valve impacts.
5. The method of claim 1 , wherein determining whether the pump is in the wear condition includes estimating the peak count over a plurality of pump cycles and calculating a trend of the peak counts.
6. The method of claim 1 , wherein determining whether the pump is in the wear condition includes:
estimating the peak count for each of a plurality of successive pump cycles;
identifying a number of pump cycles having a peak count that exceeds the expected peak count; and
determining that the pump is in the wear condition in response to the number of pump cycles exceeding a selected percentage of a total number of the plurality of successive pump cycles.
7. The method of claim 1 , wherein determining whether the pump is in the failure condition includes:
estimating the peak count for each pump cycle;
comparing the peak count to a peak count threshold value that is greater than the expected peak count;
determining that the pump is in the failure condition in response to the peak count exceeding the peak count threshold value.
8. The method of claim 1 , wherein determining whether the pump is in the wear condition and determining whether the pump is in the failure condition are performed concurrently in real time during an energy industry operation.
9. The method of claim 1 , further comprising, replacing a worn component where there is an indication of a wear condition.
10. The method of claim 1 , wherein the pump operation is automatically terminated when there is an indication of a failure condition.
11. A system for pressurizing fluid comprising:
a fluid pump comprising an internal component and a discharge in communication with a borehole;
at least one sensor that is responsive to vibrations of the pump, that is responsive to vibrations of the internal component, and that is strategically located to be sensitive to the vibrations of the internal component; and
a processor configured to perform:
receiving time domain measurement data from the at least one sensor, the measurement data indicating the vibrations occurring in the fluid pump and vibrations of the internal component;
filtering the measurement data to remove measurement data components associated with cyclical motions of the fluid pump;
dividing the filtered measurement data into a plurality of subsets, each subset corresponding to a pump cycle;
estimating a peak count for each subset, the peak count being a number of peaks having an amplitude that exceeds a selected amplitude threshold, the amplitude threshold associated with impacts of the internal components;
comparing the peak count with an expected peak count;
determining whether the internal component is in a condition selected from at least one of a wear condition and a failure condition based on the comparison; and
indicating a wear condition when it is determined that the internal component is in a wear condition, and indicating a failure condition when it is determined that the internal component is in a failure condition so that an operation of the pump is changed based on what is indicated.
12. The system of claim 11 , wherein the time domain measurement data is real time data collected during an energy industry operation, and the determining is performed in real time during the operation.
13. The system of claim 11 , wherein the pump is a positive displacement pump.
14. The system of claim 13 , wherein the internal component comprises an a valve, and wherein the at least one sensor comprises an accelerometer disposed on the valve.
15. The system of claim 13 , wherein the internal component comprises a sealing component, and wherein the at least one sensor comprises an accelerometer disposed proximate to the sealing component.
16. The system of claim 11 , wherein determining whether the internal component is in the wear condition includes:
estimating the peak count for each of a plurality of successive pump cycles;
identifying a number of pump cycles having a peak count that exceeds the expected peak count; and
determining that the interim component pump is in the wear condition in response to the number of pump cycles exceeding a selected percentage of a total number of the plurality of successive pump cycles.
17. The system of claim 16 , wherein determining whether the internal component is in the failure condition includes:
estimating the peak count for each pump cycle;
comparing the peak count to a peak count threshold value that is greater than the expected peak count;
determining that the internal component is in the failure condition in response to the peak count exceeding the peak count threshold value.
18. The system of claim 17 , wherein determining whether the internal component is in the wear condition and determining whether the internal component is in the failure condition are performed concurrently in real time during an energy industry operation.
19. The system of claim 11 , wherein the processor is configured to further perform, in response to determining that the internal component is in the wear condition, notifying a user the internal component is worn and should be replaced.
20. The system of claim 11 , wherein the processor is configured to further perform, in response to determining that the internal component is in the failure condition, performing at least one of:
notifying a user that the internal component has failed; and
automatically shutting down the pump by the processor.