Program for the detection and diagnosis of broken rotor bars in squirrel-cage motors

This paper presents the implementation and application of different methodologies for assessing the state of the squirrel-cage rotor of induction motors, using a digital embedded system programmed for this purpose. Also, an index is proposed in order to detect the early degradation of the rotor through the Wavelet Transform. The detection and evaluation of the rotor's health is carried out by a program developed for this end by the authors. By means of the user interface, the method for processing the start-up armature current is selected by the operator. The program is tested considering a series of operation conditions, including: steady and transient conditions. On the former, the classical approach with the aid of the Fast Fourier Transform (FFT) is applied, while in the latter the program utilizes both, the Short-Time Fourier Transform and the Wavelet Transform. Acceptable results are obtained using the techniques mentioned above for both, steady and transient conditions. The program outputs not just graphical results, also quantifies the level of fault by means of proposed rotor condition indexes; the results were obtained through laboratory tests using motors with different deterioration levels on the rotor.