Browsing by Author "Belahcen, Anouar, Assistant Prof., Aalto University, Department of Electrical Engineering and Automation, Finland"

Now showing 1 - 1 of 1

On the different magneto-mechanical waves in faulty induction motors
(2015) Martinez, Javier
School of Electrical Engineering | Doctoral dissertation (article-based)
This thesis deals with the modelling and characterization of different faults existing in induction motors. The faults considered in this thesis are broken bars, inter-turn short circuit and dynamic eccentricity. The reason of choosing these faults is because they are the most common source of problems for this type of motor.The thesis provides a set of both analytical and computational models. The analytical models are derived using permeance functions while the computational models are using the Finite Element method. The analytical model is used to quickly determine the set of existing magneto-mechanical waves in the air-gap of the induction motor. The computational models allow us to observe physical fields that are difficult to be tested experimentally with a high degree of accuracy. The thesis also includes a third section that explains the set of signal processing techniques applied to analyse the set of time-series of faulty induction motor. Using the former set of methods, I have been able to make the following achievements in the field of diagnosis. Firstly, I have been studying the relationship between broken bar fault and the interaction existing between saturation harmonics and torque ripples. Secondly, I have been studying the interaction between skew in induction motors and the magnitude of the saturation harmonics existing in the stator currents. Thirdly, I have been able to successfully model a motor suffering from eccentricity using Arbitrary Lagrangian Eulerian description of magnetic fields. This computational model, together with a permeance model, was able to justify the existing limitation of stator current analysis in induction motors. This limitation can lead to ambiguity diagnosis when using stator current analysis. To overcome this situation, the author suggest multidimensional spectral analysis to infer the existing waves in the air-gap. Finally, the author implements spectral estimators to improve the detection of the different faults.