Motor Current Signal Analysis Based on a Matched Subspace Detector

• Published in: IEEE transactions on instrumentation and measurement Vol. 66; no. 12; pp. 3260 - 3270
• Main Authors: Elbouchikhi, Elhoussin, Choqueuse, Vincent, Auger, Francois, El Hachemi Benbouzid, Mohamed
• Format: Journal Article
• Language: English
• Published: New York IEEE 01-12-2017; The Institute of Electrical and Electronics Engineers, Inc. (IEEE); Institute of Electrical and Electronics Engineers
• Subjects: Bearing faults; broken rotor bars; Detectors; diagnosis; eccentricity fault; Electric power; Engineering Sciences; Fault detection; induction motor; Induction motors; Interference; Likelihood ratio; matched subspace detector; motor current signature analysis (MCSA); oblique projection; Rotors; Sensors; Signal analysis; Signal and Image processing; Signature analysis; Stators; Induction motor; MCSA; Fault detec; Eccentricity fault; Oblique projection; Broken rotor bars; Bearing faults; Diagnosis; Subspace Matched Detector
• ISSN: 0018-9456; 1557-9662
• DOI: 10.1109/TIM.2017.2749858

Motor current signature analysis (MCSA) is a well-proven technique for electrical and mechanical faults detection in induction motors. The appearance of stator current components or increase of magnitude of some components at characteristic frequencies indicates a motor fault condition. In this paper, we propose a new MCSA fault detector based on a matched subspace technique. The proposed detector consists of three steps. First, the influence of the fundamental supply frequency is removed from the current signal using an interference cancellation technique based on oblique projection. Then, the fault-related frequency is estimated from the interference-free signal using the maximum likelihood principle. Finally, the fault detection is performed using a generalized likelihood ratio test. Simulation and experimental results illustrate the effectiveness of the proposed approach for eccentricity fault, bearing faults, and broken rotor bars detection.