Document Abstract
Quality control is applied in production process. The good condition of electrical machines can be obtained by using diagnostics. There are a
lot of methods that can be used for diagnostics of electrical machines. In the literature, standard methods are based on a study of electrical
signals. This project work start with a novel automated practical implementation for non contiguous rotor side broken bars detection and
diagnosis in induction motors. In this work a method for detection and diagnosis of rotor side broken bars is there based on the spectral
analysis via fast Fourier transform (FFT) and classification of the spectral response based on fuzzy controlled identifier. For the fault
diagnosis objective, two features are selected from the spectrum of the stator current, first is the amplitude of the harmonics representing the
broken bars defect 2sf (where s is the slip and f is the fundamental harmonics) and the second is the dc value. By using these obtained
parameters a fuzzy identifier will there to identify the number of broken bars. For the designing of this fuzzy identifier these two parameters
will be used as inputs where the decision about the state of rotor will be made. After the implementation of that work it will provide that this
technique will able to efficiently detect the number of broken bars at rotor side.