System and method for proactive motor wellness diagnosis based on potential mechanical faults

• US 20050021301A1
• Filed: 06/02/2004
• Published: 01/27/2005
• Est. Priority Date: 06/18/2003
• Status: Active Grant

• Current Assignee: Eaton Intelligent Power Limited (Eaton Corporation PLC.)
• Original Assignee: Eaton Corp. (Wisconsin) (Eaton Corporation PLC.)
• Inventors: Habetler, Thomas G., Dimino, Steven A., Krstic, Slobodan, Liu, Yanzhen, Nowak, Michael P., Obaid, Ramzy R.
• Granted Patent: US 7,117,125 B2
• Time in Patent Office: Days
• US Class Current: 702/185
• CPC Class Codes: G01R 31/343 in operation
• CPC Class Codes: G05B 2219/37351 Detect vibration, ultrasound
• CPC Class Codes: G05B 23/0229 knowledge based, e.g. expert systems; genetic algorithms
• CPC Class Codes: G05B 23/0245 based on a qualitative model, e.g. rule based; if-then decisions
• CPC Class Codes: G05B 23/0254 based on a quantitative model, e.g. mathematical relationships between inputs and outputs; functions: observer, Kalman filter, residual calculation, Neural Networks
• CPC Class Codes: H02H 7/0822 Integrated protection, motor control centres

Abstract

The present invention is directed to a system and method for proactively determining motor wellness arising from potential mechanical faults of the motor. A controller is configured to detect indicia of impending mechanical motor faults. The controller includes a processor configured to determine motor parameters of a given motor including a load and generate a set of baseline data for the given motor. The processor is also configured to acquire current spectrum data from the given motor during operation, map at least one from a plurality of load bins based on the load, and generate a mechanical fault signature from the current spectrum. The processor is caused to compare the mechanical fault signature to baseline data from the set of baseline data corresponding to the mapped bin and determine amplitude variances within the mechanical fault signature indicative of an impending mechanical fault prior to an actual mechanical fault occurrence.

Claims

23. A method of detecting impending mechanical faults comprising:

generating baseline data for the operating motor;

receiving current data from an operating motor;

performing at least one FFT on the current data to generate frequency spectrum data;

selecting system frequency sidebands within the frequency spectrum data;

summing the frequency spectrum data within the system frequency sidebands;

comparing the summed frequency spectrum data to a portion of the baseline data; and

determining amplitude variances within a component of a running shaft speed of the motor indicative of prospective faults due to at least one of motor misalignment and unbalance.

24. The method of claim 23 further comprising:

determining an average power of the operating motor;

determining a load of the operating motor;

mapping the average power of the operating motor to the a load bin based on the load of the operating motor; and

selecting a portion of the baseline data for comparison to the summed frequency data based on the mapped load bin.

25. The method of claim 23 further comprising performing a plurality of FFTs on the current data and average results from the plurality of FFTs to generate the frequency spectrum data.

26. The method of claim 23 further comprising comparing the frequency spectrum data to the baseline data within the sidebands of a shaft running frequency of the operating motor to generate a mechanical fault index indicative of a probability of prospective faults due to at least one of motor misalignment and unbalance.

27. The method of claim 23 further comprising performing at least one of an RSS value calculation and a polynomial estimation to compensate for energy leakage within the current data.

28. The method of claim 27 further comprising estimating an instantaneous amplitude of running shaft speed components as a function of neighboring running shaft speed components to perform polynomial estimation.

29. The method of claim 28 wherein the neighboring running shaft speed components include at least one upper running shaft speed component and at least one lower running shaft speed component.

30. The method of claim 27 further comprising performing the RSS value calculation for a plurality of frequency components in the system frequency sidebands.

32. The method of claim 23 further comprising communicating a wellness alert upon detecting amplitude variances indicative of future mechanical faults.

33. The method of claim 32 further comprising communicating the wellness alert to an operator counsel over a communications bus of a system including the operating motor.

34. A computer readable storage medium having stored thereon a computer program comprising instructions which, when executed by at least one processor, cause the at least one processor to:

receive operational current data from a motor;

perform at least two FFTs on the operational current data to generate frequency spectrum data;

average the frequency spectrum data;

generate a mechanical fault signature from frequency spectrum data;

if in a learning mode, compile a baseline from the mechanical fault index; and

if not in the learning mode;

determine a load of the motor;

map the average power of the motor to the a load bin based on the load of the operating motor;

compare a portion of the baseline corresponding to the mapped load bin to the mechanical fault signature; and

determine an impending mechanical fault before a fault occurrence.

1. A controller configured to detect indicia of impending mechanical motor faults and having a processor configured to:

determine motor parameters of a given motor including a load;

generate a set of baseline data for the given motor;

acquire current spectrum data from the given motor during operation;

map the current spectrum data to at least one of a plurality of load bins based on the load;

generate a mechanical fault signature from the current spectrum;

compare the mechanical fault signature to baseline data from the set of baseline data corresponding to the mapped bin; and

determine amplitude variances within the mechanical fault signature indicative of an impending mechanical fault prior to an actual mechanical fault occurrence.

2. The controller of claim 1 wherein the processor is further configured to isolate sidebands within the current spectrum data and sum the current spectrum data within the sidebands to generate the mechanical fault signature.

3. The controller of claim 1 wherein the processor is further configured to select the desired sidebands selected based on a system frequency of the given motor.

4. The controller of claim 1 wherein the processor is further configured to dynamically notch filter the current spectrum data to remove a fundamental frequency of the given motor.

5. The controller of claim 4 wherein the processor is further configured to track the fundamental frequency of the given motor over an operational range for use with an inverter.

6. The controller of claim 1 wherein the processor is further configured to digitize and decimate the current spectrum data to acquire a desired resolution for comparison with the baseline data.

7. The controller of claim 1 wherein the processor is further configured to receive raw current data and compute at least one fast Fourier Transform (FFT) to generate the current spectrum data.

8. The controller of claim 7 wherein the processor is further configured to average the current spectrum data to stabilize the current frequency spectrum data.

9. The controller of claim 1 wherein the processor is further configured to generate the mechanical fault signature by calculating a root mean square (RMS) of the current spectrum data.

10. The controller of claim 1 wherein the processor is further configured to determine a number of poles of the given motor.

11. The controller of claim 1 wherein the processor is further configured to acquire power data from the given motor and map the power data into the load bin to determine a monitoring state of the mapped load bin.

12. The controller of claim 11 wherein if the monitoring state of the mapped load bin is a learning mode, the processor is further configured to generate the set of baseline data by continuously averaging the mechanical fault signature of the given motor under known healthy conditions.

14. The controller of claim 1 wherein the processor is further configured to determine a noise pattern indicative of potential motor misalignment from the fault signature of the given motor by detecting amplitude variances within motor shaft running speed components within the current spectrum data.

15. The controller of claim 14 wherein the processor is further configured to communicate a proactive alert over a communications bus upon determining amplitude variances in the shaft running speed components greater than a threshold.

16. The controller of claim 15 wherein the processor is further configured to set the threshold according to one of a user selected threshold, a dynamically learned threshold, and a lookup table.

17. The controller of claim 1 wherein the processor is further configured to perform at least one of a root sum square (RSS) value calculation and a polynomial estimation to compensate for energy leakage within the current data.

18. The controller of claim 17 wherein the processor is further configured to estimate an instantaneous amplitude of components of running shaft speed of the given motor as a function of neighboring components of running shaft speed of the given motor to perform the polynomial estimation.

19. The controller of claim 17 wherein the processor is further configured to perform the RSS value calculation for a plurality of frequency components in the desired sidebands surrounding a shaft frequency of the given motor.

20. The controller of claim 1 wherein the processor is further configured to communicate preventative maintenance alerts over a communications bus to an operator display upon detecting indicia of at least one of motor misalignment and motor imbalance within the motor fault signature.

21. The controller of claim 1 wherein the processor is further configured to:

acquire power data from the given motor during operation;

generate a mechanical fault signature of the given motor from the power data; and

determine indicia of at least one of motor misalignment and motor imbalance within the motor fault signature.

22. The controller of claim 21 wherein the power data includes instantaneous power data.

35. The computer program of claim 34 wherein the processor is further caused to select sidebands within the frequency spectrum data around a system frequency of the motor and determine amplitude variances within the sidebands indicative of impending mechanical faults

36. The computer program of claim 34 wherein the processor is further caused to communicate a wellness alert upon detecting amplitude variances indicative of future mechanical faults.

37. The computer program of claim 36 wherein the processor is further caused to communicate the wellness alert to an operator counsel over a communications bus of a system including the motor.

38. The computer program of claim 34 wherein the processor is further caused to compensate for energy leakage in the current data.

39. The computer program of claim 38 wherein the processor is further caused to perform at least one of an RSS value calculation and a polynomial estimation to compensate for the energy leakage within the current data.

40. The computer program of claim 39 wherein the processor is further caused to estimate an instantaneous amplitude of running shaft speed components within the frequency spectrum data as a function of neighboring running shaft speed components to perform polynomial estimation.

41. The computer program of claim 39 wherein the processor is further caused to perform the RSS value calculation for a plurality of frequency components in a selected frequency band surrounding a running frequency of the motor.

41-1. The computer program of claim 34 wherein the processor is further caused to:

compile baseline power data;

receive operational power data from the motor;

perform at least two FFTs on the operational power data to generate power spectrum data;

average the power spectrum data;

generate a mechanical fault signature from power spectrum data;

compare a portion of the baseline power data corresponding to a mapped load bin to the averaged power data; and

determine an impending mechanical fault before a fault occurrence.