Method, system, and computer-readable medium for sensorless measurement of induction machine rotor bar pass frequency

• US 20040054485A1
• Filed: 09/18/2002
• Published: 03/18/2004
• Est. Priority Date: 09/18/2002
• Status: Active Grant

• Current Assignee: General Electric Company
• Original Assignee: General Electric Company
• Inventors: Rajiyah, Harindra, Dragomir-Daescu, Dan Michael, Al-Khalidy, Abdulrahman Abdallah, Kliman, Gerald Burt
• Granted Patent: US 6,789,028 B2
• Time in Patent Office: Days
• US Class Current: 702/76
• CPC Class Codes: G01P 3/48 by measuring frequency of generated current or voltage in general G01R23/00
• CPC Class Codes: H02P 21/16 Estimation of constants, e.g. the rotor time constant
• CPC Class Codes: H02P 23/14 Estimation or adaptation of motor parameters, e.g. rotor time constant, flux, speed, current or voltage

Abstract

A method comprises: sampling a current waveform of an induction machine to obtain sampled points; characterizing the sampled current over a selected first time window to obtain characterized points; subtracting the characterized points from respective sampled points to obtain a residual signal comprising residual points; obtaining a frequency spectrum representative of the residual signal over a selected second time window; and identifying a rotor bar pass frequency from the frequency spectrum. The method is typically carried out by a system comprising a computer, a computer-readable medium, or combinations thereof.

Claims

1. A method comprising:

(a) sampling a current waveform of an induction machine to obtain sampled points;

(b) characterizing the sampled current over a selected first time window to obtain characterized points;

(c) subtracting the characterized points from respective sampled points to obtain a residual signal comprising residual points;

(d) obtaining a frequency spectrum representative of the residual signal over a selected second time window; and

(e) identifying a rotor bar pass frequency from the frequency spectrum.

2. The method of claim 1 further comprising (f) obtaining an induction machine speed from the rotor bar pass frequency.

3. The method of claim 1 wherein sampling the current waveform occurs at a sampling frequency higher than a fundamental frequency of the induction machine.

4. The method of claim 3 wherein sampling occurs at a sampling frequency of at least about one hundred times the fundamental frequency.

5. The method of claim 1 wherein the first time window comprises a multiple of a fundamental period of the induction machine.

6. The method of claim 5 wherein characterizing comprises using neighboring sample points for averaging each of the sample points.

7. The method of claim 6 wherein characterizing comprises using a technique selected from the group consisting of Fourier series fitting, moving average filtering, and envelope detection.

8. The method of claim 1 wherein a beginning of the second time window is defined by a selected point.

9. The method of claim 8 wherein the selected point comprises a zero crossing within the first time window.

10. The method of claim 1 wherein a length of the second time window is about 3 to about 6 periods of a fundamental frequency of the induction machine.

11. The method of claim 1 wherein obtaining a frequency spectrum representative of the residual signal comprises obtaining a frequency spectrum of the residual signal.

12. The method of claim 1 wherein obtaining a frequency spectrum representative of the residual signal comprises (i) autocorrelating the residual signal to obtain an autocorrelation function of the residual signal;

and (ii) obtaining a frequency spectrum of the autocorrelation function of the residual signal, the frequency spectrum comprising a power spectral density.

13. The method of claim 1 wherein (e) comprises obtaining at least one significant side band of an approximate bar pass frequency and obtaining the rotor bar pass frequency from the at least one significant side band.

14. The method of claim 13 wherein (e) comprises obtaining at least two significant side bands of the approximate bar pass frequency, obtaining rotor bar pass frequency estimates from each of the at least two significant side bands, and averaging the rotor bar frequency estimates.

15. The method of claim 13 wherein (e) comprises obtaining at least two significant side bands of the approximate bar pass frequency, and further comprising

(f) obtaining an induction machine speed from the rotor bar pass frequency, wherein

(f) comprises obtaining induction machine speed estimates from each of the at least two significant side bands and averaging the induction machine speed estimates to obtain the induction machine speed.

16. A method for sensorless measurement of induction machine (22) speed, the method comprising:

(a) sampling a current waveform of the induction machine at a sampling frequency higher than a fundamental frequency of the induction machine to obtain sampled points;

(b) characterizing the sampled current over a selected first time window comprising a multiple of a fundamental period of the induction machine to obtain characterized points;

(c) subtracting the characterized points from respective sampled points to obtain a residual signal comprising residual points;

(d) obtaining a frequency spectrum representative of the residual signal over a selected second time window, wherein a beginning of the second time window is defined by a selected point;

(e) multiplying an approximate speed by the number of rotor bars of the machine to obtain an approximate bar pass frequency;

(f) obtaining at least one significant side band of the approximate bar pass frequency; and

(g) obtaining a rotor bar pass frequency from the at least one significant side band.

(h) obtaining an induction machine speed from the rotor bar pass frequency.

17. The method of claim 16 wherein sampling occurs at a sampling frequency of at least about one hundred times the fundamental frequency.

18. The method of claim 17 wherein characterizing comprises using neighboring sample points for averaging each of sample points.

19. The method of claim 18 wherein characterizing comprises using a technique selected from the group consisting of Fourier series fitting, moving average filtering, and envelope detection.

20. The method of claim 18 wherein the selected point comprises a beginning of the first time window.

21. The method of claim 18 wherein the selected point comprises a zero crossing within the first time window.

22. The method of claim 18 wherein a length of the second time window is about 3 to about 6 periods of a fundamental frequency of the induction machine.

23. The method of claim 16 wherein obtaining a frequency spectrum representative of the residual signal comprises obtaining a frequency spectrum of the residual signal.

24. The method of claim 16 wherein obtaining a frequency spectrum representative of the residual signal comprises (i) autocorrelating the residual signal to obtain an autocorrelation function of the residual signal;

and (ii) obtaining a frequency spectrum of the autocorrelation function of the residual signal, the frequency spectrum comprising a power spectral density.

25. The method of claim 16 wherein (f) comprises obtaining at least two significant side bands of the approximate bar pass frequency, and wherein (g) comprises obtaining rotor bar pass frequency estimates from each of the at least two significant side bands and averaging the rotor bar frequency estimates to obtain the rotor bar pass frequency.

26. The method of claim 16 wherein (f) comprises obtaining at least two significant side bands of the approximate bar pass frequency, and wherein (h) comprises obtaining induction machine speed estimates from each of the at least two significant side bands and averaging the induction machine speed estimates to obtain the induction machine speed.

27. A system comprising a computer configured for:

(a) sampling a current waveform of an induction machine to obtain sampled points;

(b) characterizing the sampled current over a selected first time window to obtain characterized points;

(c) subtracting the characterized points from respective sampled points to obtain a residual signal comprising residual points;

(d) obtaining a frequency spectrum representative of the residual signal over a selected second time window; and

(e) identifying a rotor bar pass frequency from the frequency spectrum.

28. The system of claim 27 wherein the computer is further configured for (f) obtaining an machine speed from the rotor bar pass frequency.

29. The system of claim 27 wherein the computer is configured for sampling the current waveform at a sampling frequency higher than a fundamental frequency of the induction machine.

30. The system of claim 27 wherein the first time window comprises a multiple of a fundamental period of the induction machine.

31. The system of claim 30 wherein characterizing comprises using neighboring sample points for averaging each of the sample points.

32. The system of claim 27 wherein a beginning of the second time window is defined by a selected point.

33. The system of claim 32 wherein the selected point is a zero crossing within the first time window.

34. The system of claim 27 wherein a length of the second time window is about 3 to about 6 periods of a fundamental frequency of the induction machine.

35. The system of claim 27 wherein obtaining a frequency spectrum representative of the residual signal comprises obtaining a frequency spectrum of the residual signal.

36. The system of claim 27 wherein obtaining a frequency spectrum representative of the residual signal comprises (i) autocorrelating the residual signal to obtain an autocorrelation function of the residual signal;

and (ii) obtaining a frequency spectrum of the autocorrelation function of the residual signal, the frequency spectrum comprising a power spectral density.

37. The system of claim 27 wherein (e) comprises obtaining at least one significant side band of an approximate bar pass frequency, and obtaining the rotor bar pass frequency from the at least one significant side band.

38. The system of claim 37 wherein (e) comprises obtaining at least two significant side bands of the approximate bar pass frequency, obtaining rotor bar pass frequency estimates from each of the at least two significant side bands, and averaging the rotor bar frequency estimates to obtain the rotor bar pass frequency.

39. The system of claim 37 wherein (e) comprises obtaining at least two significant side bands of the approximate bar pass frequency, and wherein the computer is further configured for (f) obtaining an induction machine speed from the rotor bar pass frequency by obtaining induction machine speed estimates from each of the at least two significant side bands and averaging the induction machine speed estimates.

40. A computer-readable medium storing computer instructions for instructing a computer, the computer instructions comprising:

(a) sampling a current waveform of an induction machine to obtain sampled points;

(b) characterizing the sampled current over a selected first time window to obtain characterized points;

(c) subtracting the characterized points from respective sampled points to obtain a residual signal comprising residual points;

(d) obtaining a frequency spectrum representative of the residual signal over a selected second time window; and

(e) identifying a rotor bar pass frequency from the frequency spectrum.

41. The medium of claim 40 wherein the instructions further comprise (f) obtaining an induction machine speed from the rotor bar pass frequency.

42. The medium of claim 40 wherein the instructions comprise sampling the current waveform at a sampling frequency higher than a fundamental frequency of the induction machine.

43. The medium of claim 40 wherein the first time window comprises a multiple of a fundamental period of the induction machine.

44. The medium of claim 43 wherein the instructions comprise characterizing by using neighboring sample points for averaging each of the sample points.

45. The medium of claim 40 wherein a beginning of the second time window is defined by a selected point.

46. The medium of claim 45 wherein the selected point comprises a zero crossing within the first time window.

47. The medium of claim 40 wherein a length of the second time window is about 3 to about 6 periods of a fundamental frequency of the induction machine.

48. The medium of claim 40 wherein the instructions comprise obtaining a frequency spectrum representative of the residual signal by obtaining a frequency spectrum of the residual signal.

49. The medium of claim 40 wherein the instructions comprise obtaining a frequency spectrum representative of the residual signal by (i) autocorrelating the residual signal to obtain an autocorrelation function of the residual signal;

and (ii) obtaining a frequency spectrum of the autocorrelation function of the residual signal, the frequency spectrum comprising a power spectral density.

50. The medium of claim 40 wherein the instructions comprise identifying the rotor bar pass frequency by obtaining at least one significant side band of an approximate bar pass frequency, and obtaining the rotor bar pass frequency from the at least one significant side band.

51. The medium of claim 50 wherein (e) comprises obtaining at least two significant side bands of the approximate bar pass frequency, obtaining rotor bar pass frequency estimates from each of the at least two significant side bands, and averaging the rotor bar frequency estimates to obtain the rotor bar pass frequency.

52. The medium of claim 50 wherein (e) comprises obtaining at least two significant side bands of the approximate bar pass frequency, and wherein the instructions further comprise (f) obtaining an induction machine speed from the rotor bar pass frequency by obtaining induction machine speed estimates from each of the at least two significant side bands and averaging -the induction machine speed estimates.