SYSTEM AND METHOD FOR PREDICTING MECHANICAL FAILURE OF A MOTOR

• US 20130013138A1
• Filed: 06/29/2012
• Published: 01/10/2013
• 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: Kumar, Ajith Kuttannair, Lu, Yinghui
• Granted Patent: US 9,845,012 B2
• Time in Patent Office: Days
• US Class Current: 701/22
• CPC Class Codes: B60L 2200/26 Rail vehicles
• CPC Class Codes: B60L 2220/42 with use of more than one motor
• CPC Class Codes: B60L 2240/421 Speed
• CPC Class Codes: B60L 2240/423 Torque
• 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 predicting mechanical failure of a traction motor in a vehicle includes monitoring first characteristics of an electrical signal supplied to a traction motor of a vehicle during a first detection window. The first characteristics represent a first motor electrical signature for the traction motor. The method also includes deriving one or more signature values from a first mathematical model of the first motor electrical signature and predicting a mechanical failure of the traction motor based on the one or more signature values.

Claims

1. A method comprising:

monitoring first characteristics of an electrical signal supplied to a traction motor of a vehicle during a first detection window, the first characteristics representing a first motor electrical signature for the traction motor;

deriving one or more signature values from a first mathematical model of the first motor electrical signature; and

predicting a mechanical failure of the traction motor based on the one or more signature values.

2. The method of claim 1, wherein the first detection window is associated with one or more designated operating conditions of the traction motor and monitoring the characteristics occurs when one or more actual operating conditions of the traction motor exceed the one or more designated operating conditions.

3. The method of claim 2, wherein the one or more designated operating conditions include at least one of a designated speed or a designated torque of the actual traction motor and the one or more actual operating conditions include at least one of an actual speed or an actual torque of the traction motor.

4. The method of claim 2, wherein monitoring the first characteristics includes recording the first characteristics when the one or more actual operating conditions of the traction motor exceed the one or more designated operating conditions and disregarding the first characteristics when the one or more actual operating conditions of the traction motor does not exceed the one or more designated operating conditions.

5. The method of claim 2, further comprising controlling the traction motor to cause the at least one of the actual speed or the actual torque to exceed the at least one of the designated speed or the designated torque associated with the first detection window.

6. The method of claim 1, wherein predicting the mechanical failure includes determining that the traction motor will mechanically fail when the one or more signature values exceed a designated threshold.

7. The method of claim 1, further comprising:

monitoring different, second characteristics of the electrical signal supplied to the traction motor, the second characteristics representing a second motor electrical signal for the traction motor;

deriving one or more signature values from a second mathematical model of the second motor electrical signature; and

combining the one or more signature values from the first mathematical model of the first motor electrical signature with the one or more signature values from the second mathematical model of the second motor electrical signal to form one or more representative signature values,

wherein predicting the mechanical failure includes comparing the one or more representative signature values to one or more designated thresholds.

8. The method of claim 7, wherein predicting the mechanical failure includes comparing a set of at least two of the representative signature values to a classification boundary and identifying the mechanical failure as likely to occur if at least one of the representative signature values in the set exceeds the classification boundary.

9. The method of claim 7, wherein combining the one or more signature values from the first mathematical model with the one or more signature values from the second mathematical model includes combining the one or more signature values from the first mathematical model with the one or more signature values from the second mathematical model to derive the one or more representative signature values.

10. The method of claim 1, wherein predicting the mechanical failure includes tracking the one or more signature values over time to determine a trend in the one or more signature values and estimating when the mechanical failure will occur based on the trend.

11. The method of claim 1, wherein monitoring the first characteristics of the electrical signal supplied to the traction motor includes sampling the first characteristics at a rate that is based on a speed at which the traction motor operates.

12. The method of claim 1, further comprising monitoring second characteristics of the electrical signal supplied to the traction motor during a different, second detection window, the first detection window and the second detection window associated with different operating conditions of the traction motor, and wherein predicting the mechanical failure includes identifying a failure mode based on which of the first characteristics or the second characteristics are used to predict the mechanical failure.

13. The method of claim 1, wherein predicting the mechanical failure includes comparing the one or more signature values to two or more designated thresholds and identifying a degree of predicted failure based on which of the two or more designated thresholds are exceeded by the one or more signature values.

14. A system comprising:

a signature generation module configured to monitor first characteristics of an electrical signal supplied to a traction motor of a vehicle during a first detection window, the first characteristics representing a first motor electrical signature for the traction motor; and

an analysis module configured to determine a first mathematical model of the first motor electrical signal and to derive one or more signature values from the first mathematical model, wherein the analysis module is further configured to predict a mechanical failure of the traction motor based on the one or more signature values.

15. The system of claim 14, wherein the first detection window is associated with one or more designated operating conditions of the traction motor and the signature generation module is configured to monitor the characteristics occurs when one or more actual operating conditions of the traction motor exceed the one or more designated operating conditions.

16. The system of claim 15, wherein the one or more designated operating conditions include at least one of a designated speed or a designated torque of the actual traction motor and the one or more actual operating conditions include at least one of an actual speed or an actual torque of the traction motor.

17. The system of claim 15, wherein the signature generation module is configured to monitor the first characteristics by recording the first characteristics when the one or more actual operating conditions of the traction motor exceed the one or more designated operating conditions, the signature generation module also configured to disregard the first characteristics when the one or more actual operating conditions of the traction motor does not exceed the one or more designated operating conditions.

18. The system of claim 15, further comprising an input/output (I/O) module configured to generate control signals for transmission to a control subsystem disposed onboard the vehicle to cause the at least one of the actual speed or the actual torque to exceed the at least one of the designated speed or the designated torque associated with the first detection window.

19. The system of claim 14, wherein the analysis module is configured to predict the mechanical failure by determining that the traction motor will mechanically fail when the one or more signature values exceed a designated threshold.

20. The system of claim 14, wherein the signature generation module is configured to monitor different, second characteristics of the electrical signal supplied to the traction motor, the second characteristics representing a second motor electrical signal for the traction motor,

wherein the analysis module is configured to derive one or more signature values from a second mathematical model of the second motor electrical signature and to combine the one or more signature values from the first mathematical model of the first motor electrical signature with the one or more signature values from the second mathematical model of the second motor electrical signal to form one or more representative signature values, and

wherein the analysis module is further configured to predict the mechanical failure by comparing the one or more representative signature values to one or more designated thresholds.

21. The system of claim 20, wherein the analysis module is configured to predict the mechanical failure by comparing a set of at least two of the representative signature values to a classification boundary and to identify the mechanical failure as likely to occur if at least one of the representative signature values in the set exceeds the classification boundary.

22. The system of claim 20, wherein the analysis module is configured to combine the one or more signature values from the first mathematical model with the one or more signature values from the second mathematical model by combining the one or more signature values from the first mathematical model with the one or more signature values from the second mathematical model to derive the one or more representative signature values.

23. The system of claim 14, wherein the analysis module is configured to predict the mechanical failure by tracking the one or more signature values over time to determine a trend in the one or more signature values and estimating when the mechanical failure will occur based on the trend.

24. The system of claim 14, wherein the signature generation module is configured to monitor the first characteristics of the electrical signal supplied to the traction motor by sampling the first characteristics at a rate that is based on a speed at which the traction motor operates.

25. The system of claim 14, wherein the signature generation module is configured to monitor second characteristics of the electrical signal supplied to the traction motor during a different, second detection window, the first detection window and the second detection window associated with different operating conditions of the traction motor, and

wherein the analysis module is configured to predict the mechanical failure by identifying a failure mode based on which of the first characteristics or the second characteristics are used to predict the mechanical failure.

26. The system of claim 14, wherein the analysis module is configured to predict the mechanical failure by comparing the one or more signature values to two or more designated thresholds and to identify a degree of predicted failure based on which of the two or more designated thresholds are exceeded by the one or more signature values.

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

generating 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, wherein each motor is a same type of motor and includes a same type of bearing as the other motors;

identifying one or more fault measures of each of the motor electrical signatures, the fault measures indicative of mechanical characteristics of the bearings of the motors;

performing a comparison of 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 a mechanical failure of the bearing of one of the motors based on the comparison.

28. The method of claim 27, wherein for each motor, data for generating the motor electrical signatures for the motor is collected only during a detection window when the motor is operating within at least one of a designated speed range or a designated torque range.

29. The method of claim 27, wherein predicting the mechanical failure of said one of the motors comprises:

recording a time series of the motor electrical signals for said one of the motors;

extrapolating the time series to determine predicted signature values for said one of the motors; and

predicting a remaining lifespan of said one of the motors based on the predicted signature values.