Fault detection based on current signature analysis for a generator

1. A method for detecting mechanical faults in an electromechanical machine having multiple electrical phases, the method comprising:

acquiring electrical signals for each phase of the multiple electrical phases from one or more sensors representative of an operating condition of the electromechanical machine;

normalizing the electrical signals to extract spectral information; and

detecting a fault based on analysis of the spectral information comprising

decomposing electrically balanced, non-fault related components of the multiple electrical phases into a DC quantity and electrically unbalanced, fault related components of the multiple electrical phases into an AC quantity to eliminate balanced components from the spectral information and extract unbalanced components that appear in the spectral information during abnormal operating conditions.

2. The method of claim 1 , wherein detecting a fault based on analyzing the spectral information comprises detecting a bearing fault or a gearbox fault.

3. The method of claim 1 , wherein the electrical signals representative of an operating condition of the electromechanical machine comprises current and voltage signals.

4. The method of claim 1 , wherein a current spectrum of the electromechanical machine is normalized based on at least one of Fast Fourier Transform, time frequency analysis, and multimodal resolution analysis.

5. The method of claim 1 , wherein eliminating balanced components from the spectral information comprises eliminating symmetrical components that compose an electromechanical machine current spectrum during a normal operating condition of the electromechanical machine.

6. The method of claim 1 , wherein detecting a fault based on analyzing the spectral information comprises detecting a fault in a drive-train.

7. The method of claim 1 , wherein detecting a fault further comprises:

squaring instantaneous values of the electrical signals for each of the multiple phases; and

summing the squared instantaneous values to transform the electrically balanced components into the DC quantity and any electrically unbalanced, fault-related components into the AC quantity to enhance detection sensitivity.

8. A method for detecting mechanical faults in an electromechanical machine, the method comprising:

acquiring electrical signals from one or more sensors representative of an operating condition of the electromechanical machine;

normalizing the electrical signals to extract spectral information;

analyzing the spectral information to form baseline electrical signals, wherein analyzing comprises

eliminating balanced components from the spectral information, and

extracting unbalanced components that appear in the spectral information of the electromechanical machine during abnormal operating conditions;

acquiring a second set of electrical signals representative of a non-faulty operating condition of the electromechanical machine;

transforming the electrical signals representative of the operating condition of the electromechanical machine into hybrid electrical signals using the second set of electrical signals;

normalizing the hybrid electrical signals to extract spectral information; and

detecting a fault if the hybrid electrical signals deviate from the baseline electrical signals.

9. The method of claim 8 , wherein acquiring a second set of electrical signals representative of a non-faulty operating condition of the electromechanical machine comprises acquiring a second set of electrical signals for each phase of multiple phases of the electromechanical machine.

10. The method of claim 8 , wherein normalizing the hybrid electrical signals to extract spectral information comprises:

dynamically eliminating symmetrical components from the hybrid electrical signals;

comparing a magnitude and phase of the asymmetrical components in the hybrid electrical signals with a magnitude and phase of the second set of electrical signals; and

detecting a fault when the asymmetrical components in the hybrid electrical signals deviate from the magnitude and phase of the second set of electrical signals.

11. The method of claim 10 , wherein detecting a fault comprises detecting a fault in a rotor of the electromechanical machine.

12. The method of claim 1 wherein analyzing the extracted feature comprises comparing an extracted feature with a determined threshold value.

13. A system for detecting a fault of an electromechanical machine having multiple electrical phases, the system comprising:

one or more sensors for acquiring electrical signals for each phase of the multiple electrical phases representative of an operating condition of the electromechanical machine;

a controller coupled to the one or more sensors to detect a fault in the electromechanical machine, the controller configured to

normalize the electrical signals to extract spectral information; and

analyze the spectral information by decomposing electrically balanced, non-fault related components into a DC quantity and electrically unbalanced, fault related components into an AC quantity.

14. The system of claim 13 , wherein said controller detects a bearing fault or a gearbox fault.

15. The system of claim 13 , wherein said one or more sensors acquire electrical signals for each phase of the multiple phases of the electromechanical machine.

16. The system of claim 13 , wherein the electromechanical machine comprises a drive-train.

17. The system of claim 13 , wherein the electromechanical machine comprises a generator.

18. The method of claim 13 , wherein the controller is further configured to

square instantaneous values of the electrical signals for each of the multiple phases; and

sum the squared instantaneous values to transform the electrically balanced components into the DC quantity and any electrically unbalanced, fault-related components into the AC quantity to enhance detection sensitivity.