System and method for predicting mechanical failure of a motor

• US 9,050,894 B2
• Filed: 07/06/2011
• Issued: 06/09/2015
• Est. Priority Date: 07/06/2011
• Status: Active Grant

• Current Assignee: GE Global Sourcing LLC (Westinghouse Air Brake Technologies Corporation)
• Original Assignee: General Electric Company
• Inventors: Banerjee, Arijit, Mukherjee, Rupam, Kumar, Ajith, Ramachandrapanicker, Somakumar, Boyanapally, Srilatha, Katta, Mohan Kumar
• Primary Examiner(s): LE, TOAN M
• Application Number:
  US13/176,828
• Publication Number:
  US 20130013231A1
• Time in Patent Office: 1,434 Days
• Field of Search: 702/58, 702/57, 702 64- 67, 702 71- 72, 702 75- 77, 702/81, 702/84, 702/127, 702182-183, 702/185, 702/189, 702/199, 310/40.R, 310/46, 310/72, 310/75.R, 310/103, 310/105, 310109-110, 310158-159, 310/166, 310/168, 310/177, 318/5, 318/15, 318/34, 318/44, 318 49- 52, 318111-114, 318/140, 318/143, 318146-148, 318/157, 318/163, 318244-246, 318249-250, 318725-727, 318/798, 318/807, 318/814, 318/825, 324/500, 324512-513, 324/525, 324/600, 324/612, 324/620, 324622-623, 700/9, 700 21- 22, 700 26- 27, 700286-287, 700/293, 700/297, 700/304
• US Class Current: 1/1
• CPC Class Codes: B60L 2200/26 Rail vehicles
• CPC Class Codes: B60L 2260/44 by parameter estimation
• CPC Class Codes: B60L 2260/50 by future state prediction
• CPC Class Codes: B60L 3/0061 relating to electrical machines
• CPC Class Codes: Y02T 10/64 Electric machine technologies in electromobility
• CPC Class Codes: Y02T 10/84 Data processing systems or methods, management, administration

Abstract

A method for a vehicle having plural electric motors includes generating motor electrical signatures for the motors. The motor electrical signatures represent one or more characteristics of electrical energy that is supplied to the motors. The method also includes identifying one or more fault measures of each of the motor electrical signatures. The fault measures are indicative of mechanical characteristics of the motors. The method further includes comparing the one or more fault measures of each motor with the one or more fault measures of the other motors in the same vehicle and predicting an impending mechanical failure of one or more of the motors based on comparing the fault measures of the motors.

Claims

1. A method for a vehicle having plural electric motors, the method comprising:

using one or more processors to generate motor electrical signatures for the motors, the motor electrical signatures representative of one or more characteristics of electrical energy that is supplied to the motors;

using the one or more processors to identify one or more disturbance peaks of the electrical signatures, wherein one or more fault measures of each of the motor electrical signatures, are based on the number of disturbance peaks and indicative of mechanical characteristics of the motors;

using the one or more processors to compare the one or more fault measures of each motor with the one or more fault measures of the other motors in the same vehicle;

using the one or more processors to predict an impending mechanical failure of one or more of the motors based on comparing the fault measures of the motors; and

using the one or more processors, responsive to the impending mechanical failure that is predicted, to generate a signal relating to at least one of vehicle control to bring the vehicle to a stop or vehicle movement for the vehicle to travel to a designated maintenance facility.

2. The method of claim 1, wherein the motor electrical signatures are frequency-domain spectra of the characteristics of the electrical energy supplied to the motors, and the one or more disturbance peaks in the frequency domain spectra are based on amplitudes of the disturbance peaks.

3. The method of claim 2, wherein the identifying step includes determining harmonic frequencies of the motor electrical signatures that are based on operating speeds of the motors and identifying the disturbance peaks in subharmonic frequency ranges of the motor electrical signatures.

4. The method of claim 1, wherein the identifying step includes determining which of a plurality of the disturbance peaks in the motor electrical signatures exceed one or more first thresholds, selecting one or more groups of the peaks that exceed the thresholds, identifying at least one of the peaks in each of the groups as a representative disturbance peak, and comparing the representative disturbance peaks to one or more second thresholds to determine if the representative disturbance peaks represent the fault measures.

5. The method of claim 1, wherein the comparing step includes, for each of the motors being examined, calculating a first average or median of the fault measures associated with the plural motors, calculating a second average or median of the fault measures associated with each of the plural motors other than the motor being examined, and comparing the first average or median with the second average or median, wherein the predicting step includes predicting the impending mechanical failure of the motor being examined based on the comparing step.

6. The method of claim 1, further comprising using the one or more processors to determine the one or more characteristics of electrical energy that is supplied to the motors, by calculating mutual impedance characteristics of the motors, and wherein the identifying step includes determining if the mutual impedance characteristics exceed one or more thresholds.

7. The method of claim 1, further comprising using the one or more processors to determine the one or more characteristics of electrical energy that is supplied to the motors, by measuring electric currents that are induced by magnetic fields generated by the motors, and wherein the identifying step includes determining if the electric currents exceed one or more thresholds.

8. The method of claim 1, wherein the generating step includes generating the motor electrical signatures for traction motors in a rail vehicle capable of self-propulsion, the comparing step includes comparing the one or more fault measures of each of the traction motors with the one or more fault measures of the other traction motors in the same rail vehicle, and the predicting step includes predicting the impending mechanical failure of one or more of the traction motors based on comparing the fault measures of the traction motors.

9. The method of claim 1, wherein the predicting step includes predicting the impending mechanical failure of one or more bearings of the one or more of the motors.

10. The method of claim 1, wherein the signal is further communicated to off-board the vehicle.

11. A failure prediction system comprising:

an input/output (I/O) module configured to determine one or more characteristics of electrical energy that is supplied to plural electric motors in a vehicle;

a signature generation module configured to create motor electrical signatures for the motors, the motor electrical signatures representative of the characteristics of the electrical energy supplied to the motors; and

an analysis module configured to identify one or more disturbance peaks and determine one or more fault measures of the motor electrical signatures of one or more of the motors based on the number of disturbance peaks, the fault measures indicative of mechanical characteristics of one or more of the motors, wherein the analysis module is configured to compare the fault measures of each motor with the fault measures of the other motors in the same vehicle, predict an impending mechanical failure of one or more of the motors based on comparing the fault measures of the motors and, responsive to the impending mechanical failure that is predicted, generate a signal relating to at least one of vehicle control to bring the vehicle to a stop or vehicle movement for the vehicle to travel to a designated maintenance facility.

12. The failure prediction system of claim 11, wherein the motor electrical signatures are frequency-domain spectra of the characteristics of the electrical energy supplied to the motors, and the one or more disturbance peaks in the frequency-domain spectra are based on amplitudes of the disturbance peaks.

13. The failure prediction system of claim 12, wherein the analysis module is configured to determine harmonic frequencies of the motor electrical signatures that are based on operating speeds of the motors and to identify the disturbance peaks in subharmonic frequency ranges of the motor electrical signatures.

14. The failure prediction system of claim 11, wherein the analysis module is configured to determine which of the disturbance peaks in the motor electrical signatures exceed one or more first thresholds, to select one or more groups of the peaks that exceed the thresholds, to identify at least one of the peaks in each of the groups as a representative disturbance peak, and to compare the representative disturbance peaks to one or more second thresholds to determine if the representative disturbance peaks represent the fault measures.

15. The failure prediction system of claim 11, wherein the analysis module is configured to, for each of the motors being examined, calculate a first average or median of the fault measures associated with the plural motors, calculate a second average or median of the fault measures associated with each of the plural motors other than the motor being examined, and compare the first average or median with the second average or median, the analysis module further configured to predict the impending mechanical failure of the motor being examined based on the comparison.

16. The failure prediction system of claim 11, wherein the analysis module is configured to calculate mutual impedance characteristics of the motors and to predict the impending mechanical failure based on whether the mutual impedance characteristics exceed one or more thresholds.

17. The failure prediction system of claim 11, wherein the I/O module is configured to receive measurements of electric currents that are induced by magnetic fields generated by the motors and the analysis module is configured to predict the impending mechanical failure based on whether the electric currents exceed one or more thresholds.

18. The failure prediction system of claim 11, wherein the signal generation module is configured to generate the motor electrical signatures for traction motors in a rail vehicle capable of self-propulsion and the analysis module is configured to predict the impending mechanical failure of one or more of the traction motors based on comparing the fault measures of the traction motors.

19. The failure prediction system of claim 11, wherein the analysis module is configured to predict the impending mechanical failure of one or more bearings of the one or more of the motors.

20. The failure prediction system of claim 11, wherein the signal is further communicated to off-board the vehicle.

21. A non-transitory computer readable storage medium for a system having a processor, the computer readable storage medium including one or more sets of instructions that direct the processor to:

generate motor electrical signatures for plural electric motors of a common vehicle, the motor electrical signatures representative of one or more characteristics of electrical energy that is supplied to the motors;

identify the number of one or more disturbance peaks to determine one or more fault measures of each of the motor electrical signatures, the fault measures indicative of mechanical characteristics of the motors;

compare the one or more fault measures of each motor with the one or more fault measures of the other motors in the same vehicle;

predict an impending mechanical failure of one or more of the motors based on comparing the fault measures of the motors; and

generate, responsive to the impending mechanical failure that is predicted, a signal relating to at least one of vehicle control to bring the vehicle to a stop or vehicle movement for the vehicle to travel to a designated maintenance facility.

22. The non-transitory computer readable storage medium of claim 21, wherein the motor electrical signatures are frequency-domain spectra of the characteristics of the electrical energy supplied to the motors, and the one or more disturbance peaks are in the frequency-domain spectra based on amplitudes of the disturbance peaks, and the fault measures are based on the number of disturbance peaks that are identified.

23. The non-transitory computer readable storage medium of claim 22, wherein the one or more sets of instructions direct the processor to determine harmonic frequencies of the motor electrical signatures that are based on operating speeds of the motors and to identify the disturbance peaks in subharmonic frequency ranges of the motor electrical signatures.

24. The non-transitory computer readable storage medium of claim 21, wherein the one or more sets of instructions direct the processor to, for each of the motors being examined:

calculate a first average or median of the fault measures associated with the plural motors;

calculate a second average or median of the fault measures associated with each of the plural motors other than the motor being examined;

compare the first average or median with the second average or median; and

predict the impending mechanical failure of the motor being examined based on the comparison.

25. The non-transitory computer readable storage medium of claim 21, wherein the one or more sets of instructions direct the processor to calculate mutual impedance characteristics of the motors and to predict the impending mechanical failure based on whether the mutual impedance characteristics exceed one or more thresholds.

26. The non-transitory computer readable storage medium of claim 21, wherein the one or more sets of instructions direct the processor to receive measurements of electric currents that are induced by magnetic fields generated by the motors and to predict the impending mechanical failure based on whether the electric currents exceed one or more thresholds.

27. A method for a vehicle having plural electric motors, the method comprising:

obtaining respective electrical signatures of the plural electric motors, wherein the electrical signatures relate to respective electrical energy supplied to the motors;

using one or more processors to perform a comparison of the electrical signature of one of the motors to the electrical signatures of the other motors in the same vehicle;

using the one or more processors to determine an impending mechanical failure of said one of the motors based on the comparison;

using the one or more processors to communicate a signal in response to determining the impending mechanical failure, wherein the signal is communicated to off-board the vehicle; and

using the one or more processors, responsive to the impending mechanical failure that is predicted, to generate a signal relating to at least one of vehicle control to bring the vehicle to a stop or vehicle movement for the vehicle to travel to a designated maintenance facility.

28. The method of claim 27, wherein the comparison is of subharmonic frequency ranges of the electrical signatures.

29. The method of claim 27, wherein the comparison is of frequency ranges of the electrical signatures, and wherein the frequency range for each electrical signature is smaller than a fundamental frequency of the electrical signature.