Method And System Of Monitoring And Prognostics

• US 20070282773A1
• Filed: 08/15/2007
• Published: 12/06/2007
• Est. Priority Date: 09/26/2005
• Status: Active Grant

• Current Assignee: Alexey Turischev, Gregory Harrison
• Original Assignee: Alexey Turischev, Gregory Harrison
• Inventors: Harrison, Gregory, Turischev, Alexey
• Granted Patent: US 7,478,013 B2
• Time in Patent Office: Days
• US Class Current: 706/25
• CPC Class Codes: G05B 23/024 Quantitative history assessment, e.g. mathematical relationships between available data; Functions therefor; Principal component analysis [PCA]; Partial least square [PLS]; Statistical classifiers, e.g. Bayesian networks, linear regression or correlation analysis; Neural networks
• CPC Class Codes: G06N 3/02 Neural networks

Abstract

A neural network learns the operating modes of a system being monitored under normal operating conditions. Anomalies can be automatically detected and learned. A control command can be issued or an alert can be issued in response thereto.

Claims

1. A method comprising:

sensing periodic time based indicia;

sensing a plurality of sets of conditions associated with operation of a selected system at spaced apart timed intervals;

generating representations of the sensed conditions;

processing the representations along with at least some of the time based indicia, with at least one neural network with the network exhibiting a plurality of learned states with each state corresponding to a respective set of sensed conditions; and

combining results of the processing, and responsive to the combining,

producing at least one of an alarm indicator, or, a control signal,

where processing includes comparing time evolution of sensed conditions with learned time evolution of operation of the selected system.

2. A method as in claim 1 where the sensed state of the sensors at respective intervals changes over time to create a trajectory of system states through time that is learned by the neural network.

3. A method as in claim 2 where the trajectory of states through time adapts in response to repeated training with new time-based training data to learn a system state trajectory that comprises a path that describes an average state trajectory within selected ranges, thereby creating a footprint in the memory of the system state trajectory through time.

4. A method as in claim 3 which includes, a learned state trajectory path able to monitor a state trajectory of a system in operation, and providing an indication of deviation from the previously learned trajectory path.

5. A method as in claim 4 which includes detecting and monitoring anomalies, reporting the anomalies to a user, and causing control actions to take place.

6. A method as in claim 5 which includes displaying the current set of system state parameters and their expected ranges as time progresses.

7. A method as in claim 6 which includes learning trajectories of aircraft in various flight patterns and indicating deviances from the expected flight plan.

8. A method as in claim 6 that includes learning expected time-based behavior of a plurality of ocean states at distributed locations and detecting anomalies in ocean rhythms that may be indicative of tsunami conditions or other unexpected ocean conditions.

9. A system for monitoring an operating device comprising:

at least one sensor for sensing a device related parameter;

at least one neural network for learning the normal time-and mode-based characteristics of the operating device, the neural network having an input port for receipt of information from a displaced computer network and an output port to transmit information to a displaced computer network, the neural network suitable for receiving and processing representations of sensor output information and time indicating indicia and creating stored internal representations thereof where the neural network automatically detects anomalies in the normal characteristics of the operating device by detecting time-based representations that deviate from the stored internal representations in accordance with at least one pre-established criterion.

10. A system as in claim 9 where each time indicating indicia comprises an entirely new learned neural network state corresponding to a time evolution of the sensed conditions.

11. A system as in claim 10 where the learned neural network state changes over time to create a system state trajectory through time that is learned by the neural network.

12. A system as in claim 11 where the system state trajectory adapts through time under repeated training on new time-based training data to learn an expected system state trajectory that forms a multidimensional system state trajectory path over time, the path defines an average state trajectory within controllable-accuracy acceptability range.

13. A system as in claim 12, where the learned state trajectory path is used to monitor the state trajectory of a system in operation and provide indication of deviation from the previously learned trajectory path.

14. A system as in claim 13 where deviations from the previously learned trajectory path are monitored by a control system and reported to a user as anomalies so that the user can take action to control the device in response thereto.

15. A system as in claim 14 that displays on a user interface system state parameters associated with a selected time and expected ranges for the state parameters as time progresses, the expected ranges changing as the system state evolves through time, the multidimensional acceptability range changing through time in parallel with the current system state parameters.

16. A system as in claim 15 wherein trajectories of aircraft having various flight patterns are learned and deviations from an expected flight plan can be indicated, the learned trajectories being of one of a group including multiple, temporally continuous, sections and legs of the approach and flight path.

17. A system as in claim 15 wherein the expected time-based behavior of an ocean state at distributed locations is learned to detect anomalies in normal ocean rhythms that may be indicative of unexpected ocean conditions.

18. A method for monitoring an operating device comprising:

sensing a device related parameter;

learning normal time-and mode-based characteristics of the operating device by receiving and processing representations of sensor output information and time indicating indicia and creating stored internal representations thereof; and

detecting anomalies in the normal characteristics of the operating device by detecting time-based representations that deviate from stored internal representations in accordance with at least one pre-established criterion.

19. The method of claim 18 wherein learning the normal characteristics of the operating device further comprises creating a system state trajectory through time from the time indicating indicia.

20. The method of claim 19 further comprising adapting the system state trajectory through time to learn an expected system state trajectory.

21. The method of claim 20 further comprising forming a multidimensional system state trajectory path over time that defines an average state trajectory within a controllable-accuracy acceptability range.

22. The method of claim 21 further comprising monitoring the state trajectory of a system in operation by a control system.

23. The method of claim 22 further comprising reporting deviations from the previously learned trajectory path to a user as anomalies so that the user can take action to control the device in response thereto.