Vibration engine monitoring neural network object monitoring
The present invention provides an aircraft engine vibration system that provides information about engine health. Embodiments of the present invention monitor for excessive vibration, monitor for bird strike, monitor for ice build up on the fan section, and monitor general engine health. An embodiment of the present invention utilizes neural network architecture for the detection of excessive vibration and ice detection build-up on the fan section of a turbo-fan engine and to monitor engine health through the high-pressure turbine section of the engine.
1. A method of monitoring vibration of an engine, the method comprising:
sensing a pattern of vibration of an engine;
determining whether magnitude of the vibration exceeds a predetermined threshold;
analyzing the pattern of vibration when the magnitude of the vibration exceeds the predetermined threshold; and
determining a cause of excessive vibration based upon analysis of the pattern of vibration.
2. The method of claim 1 , further comprising learning steady state operation of the engine.
3. The method of claim 2 , wherein learning the steady state operation establishes the predetermined threshold.
4. The method of claim 1 , further comprising, providing an alert regarding the excessive vibration.
5. The method of claim 4 , wherein providing the alert includes suggesting the cause of the excessive vibration.
6. The method of claim 5 , wherein the cause includes one of a bird strike, foreign object damage, and failure of a bearing of the engine.
7. The method of claim 1 , further comprising analyzing the pattern of vibration when the vibration does not exceed the predetermined threshold.
8. The method of claim 7 , wherein the pattern of vibration is analyzed for build-up of ice on a fan section of the engine.
9. The method of claim 8 , further comprising providing an alert regarding the build-up of ice.
10. The method of claim 7 , wherein the pattern of vibration is analyzed for an over-speed condition.
11. The method of claim 10 , further comprising providing an alert regarding the over-speed condition.
12. The method of claim 10 , further comprising recording duration of the over-speed condition.
13. The method of claim 1 , wherein sensing the vibration of the engine includes determining displacement of a shaft of the engine.
14. The method of claim 13 , wherein the predetermined threshold is around 3/1,000 inch (mils).
15. The method of claim 1 , further comprising performing a Hilbert transformation of the sensed vibration of the engine.
16. The method of claim 15 , further comprising inputting outputs of the Hilbert transformation into a neural network.
17. The method of claim 16 , wherein:
the neural network generates a plurality of neuron set space vectors; and
analyzing the pattern includes detecting a phase shift of the neuron set space vectors.
18. A system for monitoring vibration of an engine, the system comprising:
a plurality of accelerometers configured to sense a pattern of vibration of an engine; and
a neural network configured to receive a pattern of outputs from the plurality of accelerometers, the neural network including:
a first component configured to determine whether magnitude of the vibration exceeds a predetermined threshold;
a second component configured to analyze the pattern of vibration when the vibration exceeds the predetermined threshold; and
a third component configured to determine a cause of the excessive vibration based upon analysis of the pattern of vibration.
19. The system of claim 18 , wherein the neural network further includes a fourth component configured to learn steady state operation of the engine.
20. The system of claim 19 , wherein the fourth component is further configured to establish the predetermined threshold.
21. The system of claim 18 , further comprising a display unit configured to provide an alert from the neural network.
22. The system of claim 21 , wherein the display unit is further configured to provide a suggested cause of the excessive vibration.
23. The system of claim 22 , wherein the cause includes one of a bird strike, foreign object damage, and failure of a bearing of the engine.
24. The system of claim 21 , wherein the neural network further includes a fifth component configured to analyze the pattern of vibration when the vibration does not exceed the predetermined threshold.
25. The system of claim 24 , wherein the fifth component is further configured to analyze the pattern of vibration for build-up of ice on a fan section of the engine.
26. The system of claim 24 , wherein the fifth component is further configured to analyze the pattern of vibration for an over-speed condition.
27. The system of claim 26 , wherein the fifth component is further configured to record duration of the over-speed condition.
28. The system of claim 18 , wherein the plurality of accelerometers senses vibration of the engine by determining displacement of a shaft of the engine.
29. The system of claim 28 , wherein the predetermined threshold is around 3/1,000 inch (mils).
30. The system of claim 18 , further comprising a sixth component configured to perform a Hilbert transformation of the sensed vibration of the engine.
31. The system of claim 30 , wherein outputs of the sixth component are input into the neural network.
32. The system of claim 31 , wherein:
the neural network generates a plurality of neuron set space vectors; and
the second component is further configured to detect a phase shift of the neuron set space vectors.
33. An engine comprising:
a shaft having a first end and a second end;
a fan section coupled to the shaft toward the first end;
an engine core coupled to the shaft toward the second end; and
a system for monitoring vibration of the engine, the system including:
a plurality of accelerometers configured to sense a pattern of vibration of an engine; and
a neural network configured to receive a pattern of outputs from the plurality of accelerometers, the neural network including:
a first component configured to determine whether magnitude of the vibration exceeds a predetermined threshold;
a second component configured to analyze the pattern of vibration when the vibration exceeds the predetermined threshold; and
a third component configured to determine a cause of the excessive vibration based upon analysis of the pattern of vibration.
34. The engine of claim 33 , wherein the neural network further includes a fourth component configured to learn steady state operation of the engine.
35. The engine of claim 34 , wherein the fourth component is further configured to establish the predetermined threshold.
36. The engine of claim 33 , further comprising a display unit configured to provide an alert from the neural network.
37. The engine of claim 36 , wherein the display unit is further configured to provide a suggested cause of the excessive vibration.
38. The engine of claim 37 , wherein the cause includes one of a bird strike, foreign object damage, and failure of a bearing of the engine.
39. The engine of claim 36 , wherein the neural network further includes a fifth component configured to analyze the pattern of vibration when the vibration does not exceed the predetermined threshold.
40. The engine of claim 39 , wherein the fifth component is further configured to analyze the pattern of vibration for build-up of ice on a fan section of the engine.
41. The engine of claim 39 , wherein the fifth component is further configured to analyze the pattern of vibration for an over-speed condition.
42. The engine of claim 41 , wherein the fifth component is further configured to record duration of the over-speed condition.
43. The engine of claim 33 , wherein the plurality of accelerometers senses vibration of the engine by determining displacement of a shaft of the engine.
44. The engine of claim 33 , wherein the predetermined threshold is around 3/1,000 inch (mils).
45. The engine of claim 33 , further comprising a sixth component configured to perform a Hilbert transformation of the sensed vibration of the engine.
46. The engine of claim 45 , wherein outputs of the sixth component are input into the neural network.
47. The engine of claim 46 , wherein:
the neural network generates a plurality of neuron set space vectors; and
the second component is further configured to detect a phase shift of the neuron set space vectors.
48. An aircraft comprising:
a fuselage;
a pair of wings; and
at least one engine including:
a shaft having a first end and a second end;
a fan section coupled to the shaft toward the first end;
an engine core coupled to the shaft toward the second end; and
a system for monitoring vibration of the engine, the system including:
a plurality of accelerometers configured to sense a pattern of vibration of an engine; and
a neural network configured to receive a pattern of outputs from the plurality of accelerometers, the neural network including:
a first component configured to determine whether magnitude of the vibration exceeds a predetermined threshold;
a second component configured to analyze the pattern of vibration when the vibration exceeds the predetermined threshold; and
a third component configured to determine a cause of the excessive vibration based upon analysis of the pattern of vibration.
49. The aircraft of claim 48 , further comprising:
a first interface unit provided within the neural network and configured to provide the pattern of vibration; and
a second interface unit provided within the fuselage and configured to receive the pattern of vibration.
50. The aircraft of claim 49 , wherein the first and second interfaces units are further configured to provide and receive, respectively, the cause of the vibration.
51. The aircraft of claim 49 , wherein the first interface units include a transmitter and a receiver, respectively.
52. The aircraft of claim 48 , further comprising a fourth component configured to perform a Hilbert transformation of the sensed vibration of the engine.
53. The aircraft of claim 52 , wherein outputs of the fourth component are input into the neural network.
54. The aircraft of claim 53 , wherein:
the neural network generates a plurality of neuron set space vectors; and
the second component is further configured to detect a phase shift of the neuron set space vectors.
55. A computer software program product comprising:
first computer program code means for determining whether magnitude of a sensed pattern of vibration of an engine exceeds a predetermined threshold;
second computer program code means for analyzing the pattern of vibration when the magnitude of the vibration exceeds the predetermined threshold; and
third computer program code means for determining a cause of excessive vibration based upon analysis of the pattern of vibration.
56. The computer software program product of claim 55 , further comprising fourth computer program code means for learning steady state operation of the engine.
57. The computer software program product of claim 56 , wherein the fourth computer program code means establishes the predetermined threshold.
58. The computer software program product of claim 55 , further comprising fifth computer program code means for providing an alert regarding the excessive vibration.
59. The computer software program product of claim 58 , wherein the fifth computer program code means suggests the cause of the excessive vibration.
60. The computer software program product of claim 59 , wherein the cause includes one of a bird strike, foreign object damage, and failure of a bearing of the engine.
61. The computer software program product of claim 55 , further comprising sixth computer program code means for analyzing the pattern of vibration when the vibration does not exceed the predetermined threshold.
62. The computer software program product of claim 61 , wherein the pattern of vibration is analyzed for build-up of ice on a fan section of the engine.
63. The computer software program product of claim 62 , further comprising seventh computer program code means for providing an alert regarding the build-up of ice.
64. The computer software program product of claim 61 , wherein the pattern of vibration is analyzed for an over-speed condition.
65. The computer software program product of claim 64 , further comprising eighth computer program code means for providing an alert regarding the over-speed condition.
66. The computer software program product of claim 64 , further comprising ninth computer program code means for recording duration of the over-speed condition.
67. The computer software program product of claim 55 , wherein sensing the vibration of the engine includes determining displacement of a shaft of the engine.
68. The computer software program product of claim 67 , wherein the predetermined threshold is around 3/1,000 inch (mils).
69. The computer software program product of claim 55 , further comprising tenth computer program code means for performing a Hilbert transformation of the sensed vibration of the engine.
70. The computer software program product of claim 69 , wherein outputs of the Hilbert transformation are input into a neural network.
71. The computer software program product of claim 70 , wherein:
the neural network generates a plurality of neuron set space vectors; and
the second computer program code means detects a phase shift of the neuron set space vectors.