Method and apparatus for predicting winding failure using zero crossing times

• US 5,488,281 A
• Filed: 09/28/1994
• Issued: 01/30/1996
• Est. Priority Date: 09/28/1994
• Status: Expired due to Term

• Current Assignee: Allen-Bradley Company (Rockwell Automation, Inc.)
• Original Assignee: Allen-Bradley Company (Rockwell Automation, Inc.)
• Inventors: Wang, Yuan, Unsworth, Peter J.
• Primary Examiner(s): Shoop, Jr., William M.
• Assistant Examiner(s): Masih, Karen
• Application Number:
  US08/314,472
• Time in Patent Office: 489 Days
• Field of Search: 318/798-815, 318/636, 318/638, 318/476, 364/550, 364/474.12, 364/178, 388/907.5, 388/909
• US Class Current: 318/806
• CPC Class Codes: G01R 31/343 in operation
• CPC Class Codes: H02P 27/02 using supply voltage with constant frequency and variable amplitude

Abstract

A method and apparatus to be used with a motor controller for determining the degree of rotor winding failure and stator winding failure using current zero crossing times of the stator winding currents. An error signal sequence, consisting of phase angle errors between consecutive zero crossings over a sampling period, is generated and analyzed in different regions of interest in either the time or frequency domain, producing signals with varying amplitudes, the signal amplitudes together with other motor parameter measurements being used to determine the degree of rotor winding failure and stator winding failure.

Claims

1. A method to be used with a motor controller, the motor controller providing stator winding currents at a frequency of X cycles per second and indicating zero crossing times when each current is zero, a set of ideal rotor signal data, an acceptable rotor error value, and a sampling period, the motor being of a design having p poles, the method predicting rotor winding failure, the method comprising the steps of:

generating a sequence of error signals over the sampling period, each error signal indicating the period between consecutive current zero crossings, the plurality of error signals together forming an error signal spectrum;

analyzing the error signal spectrum to determine a signal amplitude of a rotor component within a region of interest, the region of interest being frequencies between K((2X/p)-n) and K(2X/p) Hz, where n is less than 2x/p and K is an integer;

determining a slip value indicating the difference in stator current frequency and the rotor frequency;

multiplying the signal amplitude by the slip value to produce a rotor signal; and

comparing the rotor signal to the ideal rotor signal data to produce a rotor error signal indicative of the degree of rotor failure.

2. The method as recited in claim 1 wherein the controller provides a stator period and the step of generating a plurality of error signals includes the steps of:

receiving consecutive zero crossing times from the motor controller;

subtracting each zero crossing time from a preceding consecutive zero crossing time to produce a plurality of difference values; and

subtracting 1/6th of the stator period from each difference value to produce the plurality of error signals.

3. The method as recited in claim 1 wherein the step of analyzing the error signal sequence includes the step of determining a signal frequency of said rotor component and the step of determining a slip value includes the steps of dividing the signal frequency by K(2X/p) to produce a frequency ratio and subtracting the frequency ratio from the integer one to produce the slip value.

4. The method as recited in claim 1 wherein the controller provides a slip value and the step of determining a slip includes the step of receiving the slip value from the controller.

5. The method as recited in claim 3 wherein the step of analyzing the error signal sequence includes the steps of passing the error signal sequence through a bandpass filter to obtain the rotor component having a frequency within the region of interest and determining the frequency and amplitude of the rotor component to ascertain the signal frequency and the signal amplitude.

6. The method as recited in claim 3 wherein the step of analyzing the error signal spectrum includes the steps of performing a fast Fourier transform on the error signal spectrum to produce an FFT spectrum and examining the FFT spectrum within the frequency region of interest to find the signal frequency and the signal amplitude.

7. The method as recited in claim 1 wherein p is 4, X is 60, n is at most 3, K is 1 and the predetermined period is approximately one second.

8. The method as recited in claim 1 further including the step of displaying the rotor error signal.

9. An apparatus to be used with a motor controller, the motor controller providing stator winding currents at a frequency of X Hz and indicating zero crossing times when each current is zero, a set of ideal rotor signal data, an acceptable rotor error value, and a sampling period, the motor being of a design having p poles, the apparatus for predicting rotor winding failure, the apparatus comprising:

a period calculator to receive consecutive zero crossing times from the controller over the sampling period, subtracts each zero crossing time from a preceding consecutive zero crossing time to produce a plurality of difference values, and subtracts 1/6th of the stator winding current cycle period from each difference value to produce a plurality of error signals, the plurality of error signals together forming an error signal sequence;

a frequency analyzer to analyze the error signal sequence to determine a signal amplitude of a rotor component within a frequency region of interest between K((2X/p)-n) and K(2X/p) Hz, where n is less than 2X/p and K is an integer;

a slip calculator to determines a slip value indicating the difference in stator winding current frequency and the rotor frequency;

a rotor calculator to multiply the signal amplitude by the slip value to produce a rotor signal; and

a comparator to compare the rotor condition signal to the rotor condition signal of the motor when it was healthy to produce a rotor error signal indicative of the degree of rotor failure.

10. The apparatus as recited in claim 9 wherein the frequency analyzer also analyzes the error signal sequence to determine a signal frequency of said rotor component and the slip calculator includes a divider to divide the signal frequency by K(2X/p) to produce a frequency ratio and a subtractor to subtract the frequency ratio from the integer one to produce the slip value.

11. The apparatus as recited in claim 10 wherein the frequency analyzer includes:

a bandpass filter having a bandpass to receive the error signal sequence and pass the rotor component having a frequency within the region of interest;

a frequency indicator that determines the signal frequency by determining the frequency of the rotor component; and

an amplitude indicator that determines the signal amplitude by determining the amplitude of the rotor component.

12. The apparatus as recited in claim 9 wherein the controller provides a slip value and the slip calculator determines the slip value by receiving the slip value from the controller.

13. The apparatus as recited in claim 9 wherein p is 4, X is 60, n is 3, K is 1, and the predetermined period is approximately one second.

14. A method to be used with a motor controller, the motor controller providing stator winding currents at a frequency of 60 Hz and indicating zero crossing times when each current is zero, a set of ideal rotor signal data, an acceptable rotor error value, and a sampling period, the motor being of a design having 4 poles, the method predicting rotor winding failure, the method comprising the steps of:

receiving consecutive zero crossing times from the motor controller;

subtracting each zero crossing time from a preceding consecutive zero crossing time to produce a plurality of difference values;

subtracting 1/6th of the stator winding current cycle period from each difference value to produce a plurality of error signals, each error signal indicating the period between consecutive current zero crossings, the plurality of error signals together forming an error signal sequence;

passing the error signal sequence through a bandpass filter having a bandpass between 27 and 30 Hz to produce a rotor component;

determining a signal frequency and a signal amplitude of the rotor component;

dividing the signal frequency by 30 to produce a frequency ratio;

subtracting the frequency ratio from the integer one to produce a slip value;

multiplying the signal amplitude by the slip value to produce a rotor condition signal;

comparing the rotor condition signal to the healthy rotor condition signal data to produce a rotor error signal; and

displaying the rotor error signal.

15. A method to be used with a motor controller, the motor controller providing three phase supply line voltages and stator winding currents at a frequency of X Hz and indicating zero crossing times when each current is zero, the controller also providing a set of stator signal data and a sampling period, the motor being of a design having p poles, the method predicting stator winding failure, the method comprising the steps of:

generating a plurality of error signals over the sampling period, each error signal indicating the time between consecutive current zero crossings, the plurality of error signals together forming an error signal sequence;

analyzing the error signal sequence to determine a harmonic amplitude of a stator component of the error signal spectrum at 2X Hz, the harmonic amplitude being the first element in a coordinate pair;

sampling supply voltages over the sampling period;

splitting the supply voltages into a positive sequence voltage and a negative sequence voltage;

dividing the negative sequence voltage by the positive sequence voltage to produce a voltage ratio, the voltage ratio being the second coordinate in the coordinate pair, the harmonic amplitude and the voltage ratio together forming a coordinate pair; and

comparing the coordinate pair to the stator signal data to produce a stator error signal.

16. The method as recited in claim 15 wherein the step of generating a plurality of stator error signals includes the steps of:

receiving consecutive zero crossing times from the motor controller;

subtracting each zero crossing time from a preceding consecutive zero crossing time to produce a plurality of difference values; and

subtracting 1/6th of the stator period from each difference value to produce the plurality of error signals.

17. The method as recited in claim 15 wherein the step of sampling the supply voltages includes the step of sampling all three of the supply line voltages consecutively, and the step of splitting the supply voltage includes the steps of:

passing each supply voltage signal through an analog to digital converter producing three digital voltage signals;

passing the digital voltage signals through a low pass filter to produce three fundamental voltage signals;

shifting the phase of two of the fundamental voltage signals to produce a positive sequence voltage signal wherein the negative sequence voltage is canceled out; and

shifting the phase of two of the fundamental voltage signals to produce a negative sequence voltage signal wherein the positive sequence voltage is canceled out.

18. The method as recited in claim 15 wherein the set of stator signal data includes data describing a series of curves on a graph, each curve representing voltage ratios plotted against harmonic amplitudes for a specific degree of stator winding fault, the degree of stator winding fault being different for each curve, and the step of comparing the coordinate pair to the stator signal data includes the steps of:

locating a coordinate point on one of the curves, the coordinate point corresponding to the coordinate pair;

determining a point curve, said point curve being the curve on which said coordinate point is located;

correlating the point curve with a specific degree of stator winding fault; and

producing the stator error signal.

19. The method as recited in claim 18 wherein the set of stator signal data also includes stator error data indicating harmonic amplitudes for each curve where the voltage ratio is zero, and the step of determining on which curve said coordinate is located includes the steps of:

determining the harmonic amplitude on said curve when the voltage ratio is zero to produce a zero amplitude signal; and

comparing the zero amplitude signal to the stator error data to produce the stator error signal.

20. An apparatus to be used with a motor controller, the motor controller providing three phase supply line voltages and stator winding currents at a frequency of X Hz and indicating zero crossing times when each current is zero, the controller also providing a set of stator signal data and a sampling period, the motor being of a design having p poles, the apparatus for predicting stator winding failure, the apparatus comprising:

a period calculator that receives consecutive zero crossing times from the controller over the sampling period, subtracts each zero crossing time from a preceding consecutive zero crossing time to produce a plurality of difference values, and subtracts 1/6th of the stator winding supply voltage cycle period from each difference value to produce a plurality of error signals the error signals together forming an error signal sequence;

a frequency analyzer to analyzes the error signal sequence to determine a harmonic amplitude of a stator component at 2X Hz, the harmonic amplitude being the first element in a coordinate pair;

a sensor for sampling supply voltages over the sampling period;

a voltage splitter that splits the supply voltages into a positive sequence voltage and a negative sequence voltage;

a voltage divider that divides the negative sequence voltage by the positive sequence voltage to produce a voltage ratio, the voltage ratio being the second coordinate in the coordinate pair, the harmonic amplitude and the voltage ratio together forming a coordinate pair;

a comparator that compares the coordinate pair to the stator signal data to produce a stator error signal; and

a display which receives and displays the stator error signal.

21. The apparatus as recited in claim 20 wherein the set of stator signal data includes data describing a series of curves on a graph, each curve representing voltage ratios plotted against harmonic amplitudes for a specific degree of stator winding fault, the degree of stator winding fault being different for each curve, and the comparator includes:

a point locator which places a coordinate point on one of the curves, the coordinate point corresponding to the coordinate pair;

a curve indicator which determines on which curve said coordinate point is located; and

an error signal generator which produces the stator error signal.

22. A method to be used with a motor controller, the motor controller providing three phase supply line voltages and stator winding currents at a frequency of X Hz and indicating zero crossing times when each current is zero, the controller also providing a set of stator signal data and a sampling period, the motor being of a design having p poles, the method predicting stator winding failure, the method comprising the steps of:

receiving consecutive zero crossing times from the motor controller;

subtracting each zero crossing time from a preceding consecutive zero crossing time to produce a plurality of difference values;

subtracting 1/6th of the stator period from each difference value to produce a plurality of error signals, the error signals together forming an error signal sequence;

passing the error signal spectrum through a bandpass filter having a bandpass at 2X Hz to produce a stator signal;

determining a harmonic amplitude of the stator signal, the harmonic amplitude being the first element in a coordinate pair;

sampling supply voltages over the sampling period;

splitting the supply voltages into a positive sequence voltage and a negative sequence voltage;

dividing the negative sequence voltage by the positive sequence voltage to produce a voltage ratio, the voltage ratio being the second coordinate in the coordinate pair, the harmonic amplitude and the voltage ratio together forming a coordinate pair; and

comparing the coordinate pair to the stator signal data to produce a stator error signal.

23. The method as recited in claim 22 wherein the set of stator signal data includes data describing a series of curves on a graph, each curve representing voltage ratios plotted against harmonic amplitudes for a specific degree of stator winding fault, the degree of stator winding fault being different for each curve, and the stator signal data also includes voltage zero data indicating harmonic amplitudes for each curve where the voltage ratio is zero, and the step of comparing the coordinate pair to the stator signal data includes the steps of:

locating a coordinate point on one of the curves, the coordinate point corresponding to the coordinate pair;

determining the harmonic amplitude on said curve when the voltage ratio is zero to produce a zero amplitude value;

comparing the zero amplitude value to the voltage zero data to determine on which curve said coordinate point is located; and

producing the stator error signal.