Model-based fault detection system for electric motors

• US 6,393,373 B1
• Filed: 02/11/1999
• Issued: 05/21/2002
• Est. Priority Date: 06/28/1996
• Status: Expired due to Term

• Current Assignee: Artesis Teknoloji Sistemleri A.S.
• Original Assignee: Ahmet Duyar, Arcelik A.S.
• Inventors: Duyar, Ahmet, Serafettinoglu, A. Hakan, Albas, Evren, Durakbasa, Osman Tugrul
• Primary Examiner(s): WACHSMAN, HAL D
• Application Number:
  US09/249,588
• Time in Patent Office: 1,195 Days
• Field of Search: 702/115, 702/33-35, 702/41, 702/42, 702/57-59, 702/64, 702/65, 702/81-84, 702/108, 702/112, 702/113, 702/116, 702/123-126, 702/142, 702/145-148, 702/179, 702/181-185, 702/189, 702/193, 702/196, 702/56, 702/FOR.135, 702/FOR.137, 702/FOR.139, 702/FOR.150, 702/FOR.151, 702/FOR.113, 702/FOR.114, 702/FOR.170, 702/FOR.171, 702/FOR-103, 702/FOR-123, 701/29-31, 701/34, 701/35, 701/59, 701/63, 701/99, 340/825.06, 340/825.16, 340/438, 340/635, 340/514, 340/515, 703/2-4, 703/6-8, 706/905, 706/913, 706/920, 700/21, 700/26, 700/23, 700/24-33, 700/39, 700/80, 700/81, 700/109, 700/110, 700/280, 700/287, 700/290, 700/292-294, 700/304, 700/69, 361/23, 361/24, 361/30, 361/31, 361/33, 318/563-565, 318/568.29, 318/490, 388/903, 388/907.5, 388/909, 731/16, 731/171-117, 731/19.R, 734/88, 734/89, 734/92, 734/95, 734/96
• US Class Current: 702/115
• CPC Class Codes: G01R 31/343 in operation

Abstract

The present invention relates to fault detection in a large population of motors of a common motor type. Initially, a plurality of motors of a common motor type is tested to obtain a first residual based on selected measured operating parameters. A second residual is then developed for each individual motor comprising the plurality of motors. If the difference between the first and second residuals exceeds a selected threshold, the motor generating the second residual is defective and removed from the plurality of motors and the threshold is adjusted to reflect the removal of the defective motor from the plurality of motors. The first residual may be used to monitor the operation of motors not included in the original plurality of motors to detect impending failure.

Claims

1. A method for screening manufacturing quality of a plurality of motors of a common motor type and for detecting mechanical faults in at least one motor of said plurality of motors, said method comprising the steps of:

selecting a plurality of motors comprising motors operating fault free and motors operating with one or more unknown faults;

measuring a plurality of operating signals for each motor in said plurality of motors;

using said plurality of operating signals, solving a first set of discrete state space equations for said plurality of motors, where said first set of discrete state space equations are in the form of;

2. The method of claim 1 wherein said measuring step further comprises the steps of:

measuring voltage, current and speed of each of said plurality of motors with a plurality of sensors;

multiplying the measured voltage (V), current (i) and speed (ω

) of each of said plurality of motors with selected invariants in accordance with the following equations;

3. The method of claim 2 further comprising the steps of:

combining the voltage measured for each of said plurality of motors to form a composite voltage value representative of said plurality of motors;

combining the current measured for each motor of said plurality of motors to form a composite current value representative of said plurality of motors;

combining the speed measured for each motor of said plurality of motors to form a composite speed value representative of said plurality of motors;

using said composite voltage value, said composite current value and said composite speed value, evaluating the following equations;

4. The method of claim 3 further comprising the step of determining whether each motor of said plurality of motors is faulty.

5. The method of claim 4 further comprising the step, in response to said determining step, of classifying the fault associated with said each motor of said plurality of motors.

6. The method of claim 3 further comprising the step of determining whether an additional motor of said common motor type is faulty, said determining step further comprising the steps of:

measuring a plurality of operating signals of said additional motor;

using said plurality of operating signals to solve said first set of discrete state space equations;

comparing said second set of discrete state space equations for said additional motor with said first set of discrete state space equations for said plurality of motors to identify a faulty motor;

measuring, when said additional motor is faulty, the voltage, current and speed of said additional motor;

using said measured voltage, current and speed of said additional motor, comparing each term of said first and second equations to each term of said first and second equations using said composite voltage value, said composite current value and said composite speed value; and

identifying at least one fault associated with said additional motor.

7. The method of claim 1 wherein said using step further comprises the step of selecting a threshold limit based on two standard deviations about the mean of the combined plurality of operating signals.

8. A method of testing a plurality of motors of a common motor type where said plurality of motors are in unknown operating condition, said method of testing comprising the steps of:

measuring voltage (V), current (I) and speed (ω

) of each of said plurality of motors;

multiplying the measured voltage, current and speed of each motor of said plurality of motors with selected invariants;

for each motor of said plurality of motors, calculating and retaining the result of the discrete state space equations;

9. The method of claim 8 further comprising the steps of:

generating a diagnostic model from said plurality of motors;

said diagnostic model including physical parameters and a standard deviation associated with each physical parameter;

said diagnostic model having a threshold limit associated with each of said physical parameters;

selecting a threshold value for selected physical parameters, said selected physical parameters including an inductance term (L di/dt), a motor resistance term (R i), a motor inertia term (J dω

/dt) and motor constants (k₁ ω

i, k₂ i² f ω

), where L, R, J and f denote the inductance, resistance, moment of inertia and friction coefficient of one of said plurality of motors; and

using said measured voltage, said current and said speed, comparing each term of a first equation;

10. The method of claim 9 further comprising displaying the results of said identifying step.

11. The method of claim 9 wherein said identifying step further comprises the steps of:

indicating an imbalanced rotor in response to a change of said L di/dt term in said first equation;

indicating a collector fault in response to a change of said Ri term in said first equation;

indicating a bearing fault in response to oscillatory variation of said L di/dt term in said first equation; and

indicating a bearing fault in response to change of both said L di/dt and said f ω

terms in said first and second equations.

12. A method of testing a plurality of motors of a common motor type where said plurality of motors are in an unknown operating condition, said method of testing comprising the steps of:

measuring selected operating parameters of each motor of said plurality of motors;

developing a base model using said measured selected operating parameters, said base model including a residual vector;

determining a solution for discrete state space equations for each motor of said plurality of motors, said discrete state space equations in the form of;

13. A model based fault detection and diagnosis method for detecting faults in a plurality of motors of a common motor type and developing diagnostic information for correction of said faults, said method comprising the steps of:

generating a modeled state representation of said plurality of motors, said modeled state representation including a threshold limit derived from said plurality of motors;

measuring operating parameters of each of said motors in said plurality of motors to develop a modeled state representation of each of said motors;

comparing said modeled state representation of said plurality of motors with said modeled state representation of each of said motors in said plurality of motors;

determining whether each of said motors is a faulty motor;

removing from said plurality of motors, in response to said determining step, said faulty motor;

changing said threshold limit of said modeled state representation of said plurality of motors in response to detection of said faulty motor; and

repeating said comparing, determining, removing and changing steps for each of said motors in said plurality of motors.

14. The method of claim 13 further comprising the steps of:

generating a diagnostic model from said plurality of motors, said diagnostic model including physical parameters and a standard deviation associated with each physical parameter, said diagnostic model having a threshold limit associated with each of said physical parameters;

measuring selected physical parameters of each motor in said plurality of motors;

for each motor, comparing each of said selected physical parameters to said diagnostic model; and

for each motor, identifying at least one fault, if at least one of said selected physical parameters exceeds said corresponding physical parameter threshold limit in said diagnostic model.