Method to detect faults in the rotor squirrel cage with low load in permanent state using DWT

Abstract

Although there are several methods that can be used, the Continuous Wavelet Transform and its discrete version have demonstrated their ability to work with these signals. This article presents a new method to help detect early faults in the rotor squirrel cage of induction motors noload and operational state, specifically in bars and rings. Using three techniques on the whole as it is the motor current signature analysis (MCSA), discrete wavelet transform (DWT) and multi-resolution analysis (MRA) with a decomposition tree reduced and apply a suitable index that determines the condition of the rotor. A brief description of the case study on which the research carried out is based is offered, which is used here successfully to generate a new alternative to determine the condition of the rotor squirrel cage. The performance of the results is get from the experimentation carried out and determined through comparison between the DWT conventional analysis and the new method, also exposing a brief comparison using the fourier transform. This new method reduces the uncertainty when performing the rotor diagnosis and improves the accuracy to differentiate the condition where it is.