Method and system for early detection of incipient faults in electric motors

• US 6,590,362 B2
• Filed: 07/29/2002
• Issued: 07/08/2003
• Est. Priority Date: 07/27/2001
• Status: Active Grant

• Current Assignee: Texas A&M University System
• Original Assignee: Texas A&M University System
• Inventors: Kim, Kyusung, Parlos, Alexander G
• Primary Examiner(s): Dang, Khanh
• Application Number:
  US10/207,105
• Publication Number:
  US 20030067277A1
• Time in Patent Office: 344 Days
• Field of Search: 318/798-802, 318/807-811, 702/115, 702/185
• US Class Current: 318/799
• CPC Class Codes: G01R 31/343 in operation
• CPC Class Codes: H02P 29/02 Providing protection against overload without automatic interruption of supply protection against faults of stepper motors H02P8/36

Abstract

A method and system for early detection of incipient faults in an electric motor are disclosed. First, current and voltage values for one or more phases of the electric motor are measured during motor operations. A set of current predictions is then determined via a neural network-based current predictor based on the measured voltage values and an estimate of motor speed values of the electric motor. Next, a set of residuals is generated by combining the set of current predictions with the measured current values. A set of fault indicators is subsequently computed from the set of residuals and the measured current values. Finally, a determination is made as to whether or not there is an incipient electrical, mechanical, and/or electromechanical fault occurring based on the comparison result of the set of fault indicators and a set of predetermined baseline values.

Claims

1. A method for providing an early detection of incipient electrical, mechanical and/or electromechanical faults in an electric motor, said method comprising:

acquiring current and voltage values for one or more phases of said electric motor during operation;

determining motor speed values of said electric motor during said operation;

determining a plurality of current predictions based on said voltage and said motor speed values via a neural network-based current predictor;

generating a plurality of residuals by combining said plurality of current predictions and said current values;

extracting a plurality of fault indicators from said plurality of residuals; and

determining whether or not an incipient electrical, mechanical or electromechanical fault is occurring by comparing said plurality of fault indicators with a set of predetermined baseline values.

2. The method of claim 1, wherein said extracting step further includes extracting a plurality of fault indicators using a Fast-Fourier Transform based processing technique.

3. The method of claim 1, wherein said extracting step further includes extracting a plurality of fault indicators using a wavelet-based processing technique.

4. The method of claim 1, further includes a step of generating an initial set of fault indicators as said set of predetermined baseline values.

5. The method of claim 1, further includes a step of segmenting stationary components from said acquired current and voltage values for said electric motor operating in a quasi-stationary mode.

6. A computer program product residing on a computer usable medium for providing an early detection of incipient electrical, mechanical and/or electromechanical faults in an electric motor, said computer program product comprising:

program code means for acquiring current and voltage values for one or more phases of said electric motor during operation;

program code means for determining motor speed values of said electric motor during said operation;

program code means for determining a plurality of current predictions based on said voltage and said motor speed values via a neural network-based current predictor;

program code means for generating a plurality of residuals by combining said plurality of current predictions and said current values;

program code means for extracting a plurality of fault indicators from said plurality of residuals; and

program code means for determining whether or not an incipient electrical, mechanical or electromechanical fault is occurring by comparing said plurality of fault indicators with a set of predetermined baseline values.

7. The computer program product of claim 6, wherein said program code means for extracting further includes program code means for extracting a plurality of fault indicators using a Fast-Fourier Transform-based processing technique.

8. The computer program product of claim 6, wherein said program code means for extracting further includes program code means for extracting a plurality of fault indicators using a wavelet-based processing technique.

9. The computer program product of claim 6, further includes a program code means for generating an initial set of fault indicators as said set of predetermined baseline values.

10. The computer program product of claim 6, further includes a program code means for segmenting stationary components from said acquired current and voltage values for said electric motor operating in a quasi-stationary mode.

11. A computer system for providing an early detection of incipient electrical, mechanical and/or electromechanical faults in an electric motor, said computer program product comprising:

a plurality of sensors for acquiring current and voltage values for one or more phases of said electric motor during operation;

means for determining motor speed values of said electric motor during said operation;

a neural network-based current predictor for determining a plurality of current predictions based on said voltage and said motor speed values;

means for generating a plurality of residuals by combining said plurality of current predictions and said current values;

means for extracting a plurality of fault indicators from said plurality of residuals; and

means for determining whether or not an incipient electrical, mechanical or electromechanical fault is occurring by comparing said plurality of fault indicators with a set of predetermined baseline values.

12. The computer system of claim 11, wherein said means for extracting further includes means for extracting a plurality of fault indicators using a Fast-Fourier Transform-based processing technique.

13. The computer system of claim 11, wherein said means for extracting further includes means for extracting a plurality of fault indicators using a wavelet-based processing technique.

14. The computer system of claim 11, further includes a means for generating an initial set of fault indicators as said set of predetermined baseline values.

15. The computer system of claim 11, further includes a means for segmenting stationary components from said acquired current and voltage values for said electric motor operating in a quasi-stationary mode.