Prognostication and amelioration of gas ingress in hydraulic systems

1. A method, comprising:

identifying a stable reading window when operating a carrying system comprising a hydraulic system;

collecting a set of fluid level readings of a hydraulic fluid in the hydraulic system during a plurality of stable reading windows across a corresponding plurality of operational instances of the carrying system;

calculating median fluid levels in each stable reading window of the plurality of stable reading windows;

identifying an increasing trend in the median fluid levels when analyzed as time series data that satisfies a trend threshold of at least a predefined number of successive indications of increased fluid levels in the hydraulic fluid within an analysis window; and

generating an alert of gas ingress to the hydraulic system,

wherein the alert causes injection of additional hydraulic fluid into the hydraulic system, and

wherein generating the alert of gas ingress to the hydraulic system comprises identifying that the increasing trend holds true for a majority of the predefined number of successive indications of the increased fluid levels in the hydraulic fluid.

2. The method of claim 1 , further comprising:

calculating standard deviations in the set of fluid level readings for each stable reading window of the plurality of stable reading windows;

fitting a regression model to the median fluid levels as the time series data;

calculating a mean square error for the standard deviations; and

excluding a given instance of the median fluid levels when analyzing the time series data when the mean square error associated with the given instance is above an outlier threshold.

3. The method of claim 1 , wherein:

the carrying system is an aircraft;

the stable reading window occurs a first threshold number of minutes before a middle of cruise time and the first threshold number of minutes after the middle of cruise time; and

the middle of cruise time occurs at least a second threshold number of minutes after reaching a top of climb and at least the second threshold number of minutes before reaching a top of descent during a given operational instance of the carrying system.

4. The method of claim 1 , further comprising:

identifying a given operational instance of the corresponding plurality of operational instances in which sensors in the carrying system unrelated to the hydraulic system indicate an operational aberration;

excluding a given instance of the median fluid levels associated with the given operational instance when analyzing the time series data; and

bifurcating the time series data to represent operational instances before the operational aberration and operational instances after the operational aberration.

5. The method of claim 1 , wherein analyzing the median fluid levels as the time series data further comprises:

identifying a set of successive operational instances from the corresponding plurality of operational instances;

clustering, via an unsupervised machine learning technique, the set of successive operational instances into a first cluster and a second cluster based on associated median fluid levels associated with the set of successive operational instances and associated operational times associated with the set of successive operational instances;

determining a distance between a first centroid value for the first cluster and a second centroid value for the second cluster based on the associated median fluid levels; and

excluding a latest operational instance of the set of successive operational instances from analysis in response to at least one of:

a first sum of squared distances, between the first centroid value and each member of the first cluster exceeding, a sum squared error threshold;

a second sum of squared distances, between the second centroid value and each member of the second cluster, exceeding the sum squared error threshold; and

the distance between the first centroid value and the second centroid value exceeds a service specification indicative of the hydraulic fluid being serviced within the set of successive operational instances.

6. The method of claim 1 , wherein generating the alert of gas ingress to the hydraulic system further comprises:

identifying that the increased fluid levels in the hydraulic fluid comprise at least a threshold amount over a nominal fill level for the hydraulic fluid.

7. The method of claim 1 , wherein the carrying system is a vehicle associated with at least one of:

a watercraft,

a submersible,

an aircraft,

a spacecraft,

a dirigible,

an automobile,

a train, or

an autonomous vehicle.

8. A system, comprising:

a processor; and

a memory, including instructions that when executed by the processor perform operations comprising:

identifying a stable reading window when operating a carrying system comprising a hydraulic system;

collecting a set of fluid level readings of a hydraulic fluid in the hydraulic system during a plurality of stable reading windows across a corresponding plurality of operational instances of the carrying system;

calculating median fluid levels in each stable reading window of the plurality of stable reading windows;

identifying an increasing trend in the median fluid levels when analyzed as time series data that satisfies a trend threshold of at least a predefined number of successive indications of increased fluid levels in the hydraulic fluid within an analysis window; and generating an alert of gas ingress to the hydraulic system,

wherein the alert causes injection of additional hydraulic fluid into the hydraulic system, and

wherein generating the alert of gas ingress to the hydraulic system comprises identifying that the increasing trend holds true for a majority of the predefined number of successive indications of the increased fluid levels in the hydraulic fluid.

9. The system of claim 8 , wherein the operations further comprise:

calculating standard deviations in the set of fluid level readings for each stable reading window of the plurality of stable reading windows;

fitting a regression model to the median fluid levels as the time series data;

calculating a mean square error for the standard deviations; and

excluding a given instance of the median fluid levels when analyzing the time series data when the mean square error associated with the given instance is above an outlier threshold.

10. The system of claim 8 , wherein:

the carrying system is an aircraft;

the stable reading window occurs a first threshold number of minutes before a middle of cruise time and the first threshold number of minutes after the middle of cruise time; and

the middle of cruise time occurs at least a second threshold number of minutes after reaching a top of climb and at least the second threshold number of minutes before reaching a top of descent during a given operational instance of the carrying system.

11. The system of claim 8 , wherein the operations further comprise:

identifying a given operational instance of the corresponding plurality of operational instances in which sensors in the carrying system unrelated to the hydraulic system indicate an operational aberration;

excluding a given instance of the median fluid levels associated with the given operational instance when analyzing the time series data; and

bifurcating the time series data to represent operational instances before the operational aberration and operational instances after the operational aberration.

12. The system of claim 8 , wherein analyzing the median fluid levels as the time series data further comprises:

identifying a set of successive operational instances from the corresponding plurality of operational instances;

clustering, via an unsupervised machine learning technique, the set of successive operational instances into a first cluster and a second cluster based on associated median fluid levels associated with the set of successive operational instances and associated operational times associated with the set of successive operational instances;

determining a distance between a first centroid value for the first cluster and a second centroid value for the second cluster based on the associated median fluid levels; and

excluding a latest operational instance of the set of successive operational instances from analysis in response to at least one of:

a first sum of squared distances, between the first centroid value and each member of the first cluster exceeding, a sum squared error threshold;

a second sum of squared distances, between the second centroid value and each member of the second cluster, exceeding the sum squared error threshold; and

the distance between the first centroid value and the second centroid value exceeds a service specification indicative of the hydraulic fluid being serviced within the set of successive operational instances.

13. The system of claim 8 , wherein generating the alert of gas ingress to the hydraulic system further comprises:

identifying that the increased fluid levels in the hydraulic fluid comprise at least a threshold amount over a nominal fill level for the hydraulic fluid.

14. A memory device including instructions that when executed by a processor perform operations comprising:

identifying a stable reading window when operating a carrying system comprising a hydraulic system;

collecting a set of fluid level readings of a hydraulic fluid in the hydraulic system during a plurality of stable reading windows across a corresponding plurality of operational instances of the carrying system;

calculating median fluid levels in each stable reading window of the plurality of stable reading windows;

identifying an increasing trend in the median fluid levels when analyzed as time series data that satisfies a trend threshold of at least a predefined number of successive indications of increased fluid levels in the hydraulic fluid within an analysis window; and

generating an alert of gas ingress to the hydraulic system,

wherein the alert causes injection of additional hydraulic fluid into the hydraulic system, and

wherein generating the alert of gas ingress to the hydraulic system comprises identifying that the increasing trend holds true for a majority of the predefined number of successive indications of the increased fluid levels in the hydraulic fluid.

15. The memory device of claim 14 , wherein the operations further comprise:

calculating standard deviations in the set of fluid level readings for each stable reading window of the plurality of stable reading windows;

fitting a regression model to the median fluid levels as the time series data;

calculating a mean square error for the standard deviations; and

excluding a given instance of the median fluid levels when analyzing the time series data when the mean square error associated with the given instance is above an outlier threshold.

16. The memory device of claim 14 , wherein:

the carrying system is an aircraft;

the stable reading window occurs a first threshold number of minutes before a middle of cruise time and the first threshold number of minutes after the middle of cruise time; and

the middle of cruise time occurs at least a second threshold number of minutes after reaching a top of climb and at least the second threshold number of minutes before reaching a top of descent during a given operational instance of the carrying system.

17. The memory device of claim 14 , further comprising:

identifying a given operational instance of the corresponding plurality of operational instances in which sensors in the carrying system unrelated to the hydraulic system indicate an operational aberration;

excluding a given instance of the median fluid levels associated with the given operational instance when analyzing the time series data; and

bifurcating the time series data to represent operational instances before the operational aberration and operational instances after the operational aberration.

18. The memory device of claim 14 , wherein generating the alert of gas ingress to the hydraulic system further comprises:

identifying that the increased fluid levels in the hydraulic fluid comprise at least a threshold amount over a nominal fill level for the hydraulic fluid.

19. The memory device of claim 14 , wherein the carrying system is associated with a land-based system included in a structure related to at least a dam lock.

20. The memory device of claim 14 , wherein the carrying system is a vehicle associated with at least one of:

a watercraft,

a submersible,

an aircraft,

a spacecraft,

a dirigible,

an automobile,

a train, or

an autonomous vehicle.