Different Faults and Their Diagnosis Techniques in Three-Phase Squirrel-Cage Induction Motors—A Review

Abstract

Statistics show that 80% of faults are due to eccentricity in induction motors, promoting many research efforts still devoted to the eccentricity in these motors. This paper classifies different common faults in the stator and rotor. It then presents analysis techniques for faulty induction motors. Four more applied methods in the diagnosis and detection of faults in induction motors are investigated and critiqued. The finite difference method is introduced, and its drawbacks are noted. An equivalent magnetic circuit method is then discussed as a fast but less accurate method. The reduced accuracy of the latter method is due to neglect of the spatial harmonics, the relationship between these harmonics, and time harmonics, and particularly non-linearity of the magnetic materials. An analytical method of winding functions is investigated for the analysis of faulty induction motors. This method also neglects the spatial harmonics and their relationship with time harmonics, which could considerably influence the instant of the fault event. Therefore the finite element method (FEM) is used as a powerful numerical method for analysis of faulty induction motors. However, all the previously methods have the two following drawbacks:

I.	There is no modeling algorithm that compensates the base fault error data.
II.	There is no algorithm that could diagnose and distinguish between the levels of faults.

Therefore, a combination of FEM and other methods (such as state-space, which can compensate the above-mentioned drawbacks) can be used as a precise, flexible and widely applicable technique for fault diagnosis of faulty induction motors. It is shown that this combination is implemented as an iterative process for prediction and analysis of induction motors. It concludes with some suggestions for future study.

Keywords:

• induction motors
• fault diagnosis
• finite element method