Running State Monitoring of Induction Motor Windings Using Near Infra-red Sensor Residual Signal and Q Factor Analysis

• Published in: Journal of electrical engineering & technology 2022, 17(3), , pp.1761-1774
• Main Authors: Gani, M. Ismail, Jothi Swaroopan, N. M., Shanker, N. R.
• Format: Article
• Language: English
• Subjects: Electrical Engineering; Electrical Machines and Networks; Electronics and Microelectronics; Engineering; Instrumentation; Original Article; Power Electronics; 전기공학
• ISSN: 1975-0102; 2093-7423
• DOI: 10.1007/s42835-022-01004-7

In Electric motors, identification of insulation and winding faults in stator and rotor during running state is a challenging task. Winding and insulation fault is identified through burning smell of coil, evaluating the efficiency of motor, or dismantling of motor. Motor running with winding and insulation faults lead to coil-to-coil and phase-to-phase short circuit fault. Winding insulation and winding coil fault in motor leads to unbalanced and differential flux radiation. Monitoring the winding and insulation during running state of motor is a challenging task. In this paper, monitoring of stator and rotor winding is proposed through NIR sensor during running state of motor. Near Infra-Red (NIR) sensor is fixed in air gaps of motor. NIR reflect rays from winding flux through air gaps are analysed for faults in stator and rotor winding and insulation. NRI reflected signals process with spectral band separation and NIR reflected residual (NRR) signals are obtained. NRR signal process with Tunable Q Wavelet Transform (TQWT) for monitoring and detecting, the insulation and winding fault of motor. Motor allowed to operate at different induced faults such as no load, loaded, stator, rotor insulation fault and stator, rotor-winding fault and NRR signal obtained. Q-factor base Energy band of NRR signals are analysed for winding and insulation faults through sub band energy variations. The low and high frequency component of faulty NRR signal detect with TQWT more accurately. The performance of NIR sensor-based winding and insulation fault diagnosis is compared with conventional transducers such as current signatures and radar signals. The NIR sensor based NRR signals classifies insulation and winding fault accurately of about 92% compared to current signal signatures.