Quality check of induction motor (IM) is a crucial phase for manufacturers. Among the large number of tests usually applied at the end of the production line, the vibration analysis allows to identify mechanical failures such as damaged bearings or rotor faults. This study proposes an end-of-line quality control procedure based on vibrational analysis, in both radial and axial directions, to identify mechanically faulty IMs. Many features in time and frequency domains have been extracted from time experimentally recorded signals and frequency spectra, and a suitably selected set of them has been fed into four classification models based on Logistic Regression and k-NN techniques. Results are encouraging, and an accuracy higher than 94% is achieved by all classifiers. A more extensive experimental acquisition is planned in the near future to validate the obtained results.
Mechanical fault detection for induction motors based on vibration analysis: A case study
Boria S.;Corradini M. L.;
2021-01-01
Abstract
Quality check of induction motor (IM) is a crucial phase for manufacturers. Among the large number of tests usually applied at the end of the production line, the vibration analysis allows to identify mechanical failures such as damaged bearings or rotor faults. This study proposes an end-of-line quality control procedure based on vibrational analysis, in both radial and axial directions, to identify mechanically faulty IMs. Many features in time and frequency domains have been extracted from time experimentally recorded signals and frequency spectra, and a suitably selected set of them has been fed into four classification models based on Logistic Regression and k-NN techniques. Results are encouraging, and an accuracy higher than 94% is achieved by all classifiers. A more extensive experimental acquisition is planned in the near future to validate the obtained results.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
