Three-States Fault Detection in Rolling Bearings of Induction Motors through the analysis of Stray Flux signals using the DWT

Induction motors (IMs) are relevant devices in the industry and for the different production processes they are involved. Despite being very robust machines, IMs are prone to suffer from different failures in its components. One of the most common failures in IM are those related to bearings, especially those caused by electrical erosion. In order to study this failure, most of methodologies reported in the literature induce this fault as a hole in the outer race; however, there is not a standard in the diameter of the hole, or there is no an in-depth study on the detection of the effects of the failure for different diameters. This is mainly due to the difficulty to detect it (at least with electrical analysis), since very small magnitudes are usually obtained. This paper presents a methodology which analyzes 3 states of a metallic bearing: a healthy state, light damage (3 mm) and severe damage (5 mm) in the outer race of the bearing. The proposed method relies on stray flux signals; which are associated with the kinematic chain where the bearing is located. It is used the Discrete Wavelet Transform (DWT) in order to generate a reliable index that allows to classify the 3 bearing states. The results obtained are promising and demonstrate the effectiveness of the proposal. It is validated with a 1HP IM for power supply frequencies of 15 Hz, 50 Hz and 60Hz, and for three different states.