This paper presents an artificial neural network (ANN) based technique to identify rotor faults in a three-phase induction motor. The main types of faults considered are broken bar and dynamic eccentricity. The feature extraction based on the frequency and the magnitude of the related fault components in the stator current spectrum is performed automatically by a Matlab script. Features with different speed and load levels are used as input for training a feedforward layered neural network. The laboratory results show that the proposed method is able to detect the faulty conditions with high accuracy and to separate between deferent types of faults.

Misalignment is one of the most common mechanical faults in electrical rotating machinery, it can lead to partial or total breakdown of a motor in the long run may. This paper investigates the application of the Autoregressive Model of torque signal to detect and diagnose the misalignment fault. First, the torque signal obtained from experiment in different conditions: healthy condition, angular misalignment, and parallel misalignment fault are normalized to center the signal at zero mean and scales it to unit standard deviation. Then the AR model coefficients are extracted as features. Min Redundancy max Relevance (mRMR) feature selection technique is used to select the optimal features and reduce the dimension data. Finally, the Self Organizing Map neural network is used for classification of the different conditions. The experimental results show the effectiveness of the proposed method, were both angular and parallel misalignment might be easily detected and discriminated from each other.

This paper presents a robust and reliable offline method for induction machine's rotor fault diagnosis, under low load conditions. This method is based on the analysis in frequency domain of the stator current envelope extracted by Hilbert transform. The proposed method, which is one of various demodulation techniques, is theoretically presented then experimentally validated by testing three different induction motors: one in a healthy condition, the second with one broken bar and the third with two broken bars. Each motor is subjected to three load conditions and supplied by voltage inverter which operates at different frequencies. The main results obtained by spectrum analysis of the stator current envelope are compared with these of the traditional MCSA method. The proposed method can successfully detect the fault under various operational situations, even for the lightly load, which makes it more suitable for induction motor fault diagnosis.

This paper deals with the diagnosis of electrical defects of Squirrel-Cage Rotor Induction Machines (IMs). The failures of induction machine and the diagnosis methods are presented. Among the methods, Motor Current Signature Analysis (MCSA) is used in the experimental study to detect broken rotor bars and end-ring segment. The load level and the load effects on the diagnosis are also studied. Support vector machines (SVM) is applied to classify faults.