TY - GEN
T1 - On line trained fuzzy logic and adaptive continuous wavelet transform based high precision fault detection of IM with broken rotor bars
AU - Saghafinia, A.
AU - Kahourzade, S.
AU - Mahmoudi, A.
AU - Hew, W. P.
AU - Uddin, M. Nasir
PY - 2012/12/1
Y1 - 2012/12/1
N2 - This paper presents an online trained fuzzy logic and adaptive wavelet based high precision fault detection of broken rotor bars for squirrel cage induction motor (IM). Motor faults which consist of broken rotor bars, bearing decay, eccentricity, etc. appears as different frequencies in the stator current signals. The winding function is used to obtain stator current and speed signals at different fault and load conditions. These signals are analysed through the adaptive continuous wavelet transform (CWT) to detect the amplitudes and frequency components corresponding to different fault and load conditions. The coefficients of CWT are adapted online based on the harmonics amplitude, which are the output of CWT. These amplitudes and frequencies are applied to train a fuzzy logic controller (FLC) in simulation. Then the adaptive CWT and trained FLC are applied to detect the fault condition of a large size motor in both simulation and realtime. The experimental results found that the proposed adaptive CWT and FLC based fault detection method can detect the motor fault conditions accurately. Thus, the proposed method could be a potential candidate to detect the motor fault, especially for large size industrial motors.
AB - This paper presents an online trained fuzzy logic and adaptive wavelet based high precision fault detection of broken rotor bars for squirrel cage induction motor (IM). Motor faults which consist of broken rotor bars, bearing decay, eccentricity, etc. appears as different frequencies in the stator current signals. The winding function is used to obtain stator current and speed signals at different fault and load conditions. These signals are analysed through the adaptive continuous wavelet transform (CWT) to detect the amplitudes and frequency components corresponding to different fault and load conditions. The coefficients of CWT are adapted online based on the harmonics amplitude, which are the output of CWT. These amplitudes and frequencies are applied to train a fuzzy logic controller (FLC) in simulation. Then the adaptive CWT and trained FLC are applied to detect the fault condition of a large size motor in both simulation and realtime. The experimental results found that the proposed adaptive CWT and FLC based fault detection method can detect the motor fault conditions accurately. Thus, the proposed method could be a potential candidate to detect the motor fault, especially for large size industrial motors.
KW - adaptive continuous wavelet transform
KW - broken rotor bars
KW - fault detection
KW - fuzzy logic controller
KW - squirrel cage induction motor
UR - http://www.scopus.com/inward/record.url?scp=84871660214&partnerID=8YFLogxK
U2 - 10.1109/IAS.2012.6374027
DO - 10.1109/IAS.2012.6374027
M3 - Conference contribution
AN - SCOPUS:84871660214
SN - 9781467303309
T3 - Conference Record - IAS Annual Meeting (IEEE Industry Applications Society)
BT - 2012 IEEE Industry Applications Society Annual Meeting, IAS 2012
T2 - 2012 IEEE Industry Applications Society Annual Meeting, IAS 2012
Y2 - 7 October 2012 through 11 October 2012
ER -