Classificação de diferentes métodos de filtragem aplicados ao diagnóstico de falhas de estator em motores de indução trifásicos acionados por inversores de frequência

Abstract

This course conclusion work proposes a study related to performance comparison of different filtering methods applied to the diagnosis of stator faults in three-phase induction motors driven by frequency inverters. The paper also proposes the use of the following filtering methods (i) Kernel, (ii) Principal Components Analysis, (I) Wavelet. In the classification stage the following classifiers will be used: Random Forest, Radial Basis Function (RBF Network) and k-Nearest Neighbors (k-NN). Analysis of each of the filters were carried out within each of the classifiers, in order to better compare results and, consequently, make a more effective assessment.