Title :
Back propagation neural network for classification of induction machine faults
Author :
Medoued, A. ; Lebaroud, A. ; Boukadoum, A. ; Boukra, T. ; Clerc, G.
Author_Institution :
Dept. of Electr. Eng., Univ. of 20 August 1955, Skikda, Algeria
Abstract :
This paper presents a new method for the classification of induction machine faults. The method is composed of two steps: feature extraction and classification. Feature extraction is based on the time-frequency representation (TFR), which is deliberately designed for maximizing the separability between classes. A distinct TFR is designed for each class. The classifier is designed with an artificial neural network. The flexibility of this method allows an accurate classification independent from the level of load. This method is validated on a 5.5-kW induction motor test bench.
Keywords :
backpropagation; electric machine analysis computing; fault diagnosis; feature extraction; induction motors; neural nets; pattern classification; time-frequency analysis; TFR; artificial neural network; back propagation; classifier; feature classification; feature extraction; induction machine fault classification; induction motor test bench; power 5.5 kW; time-frequency representation; Feature extraction; Jacobian matrices; Kernel; Neurons; Stators; Time frequency analysis; Training; Ambiguity Plane; Classification-Optimal TFR; Fisher´s Discriminated Ratio; Induction Machine Diagnosis; Time-Frequency; artificial neural network ANN;
Conference_Titel :
Diagnostics for Electric Machines, Power Electronics & Drives (SDEMPED), 2011 IEEE International Symposium on
Conference_Location :
Bologna
Print_ISBN :
978-1-4244-9301-2
Electronic_ISBN :
978-1-4244-9302-9
DOI :
10.1109/DEMPED.2011.6063673
