Authors:
J. F. Ramahaleomiarantsoa
1
;
N. Heraud
1
;
E. J. R. Sambatra
2
and
J. M. Razafimahenina
3
Affiliations:
1
Université de Corse, France
;
2
Institut Supérieur de Technologie, Madagascar
;
3
Ecole Supérieure Polytechnique, Madagascar
Keyword(s):
Diagnosis, Principal components analysis, Residues, Wound rotor induction machine, Modeling.
Related
Ontology
Subjects/Areas/Topics:
Informatics in Control, Automation and Robotics
;
Intelligent Control Systems and Optimization
;
Intelligent Fault Detection and Identification
Abstract:
Electrical machines are found in many applications, especially in wind energy conversion chain (WECC). However, these machines still remain the most potential of failures. Many researches and improvements have been carried out but in the aim of optimal operation systems, monitoring and diagnosis techniques are among the interests of existing laboratories and research teams. This paper deals with the principal components analysis (PCA) method application in electrical machines, especially a wound rotor induction machine (WRIM), diagnosis. The used PCA approach is based on residues analysis. To perform the matrix data needed for PCA method data input, an accurate analytical method of the WRIM is proposed. WRIM and PCA models are implemented in Matlab software. The simulation results show the potential necessity of the considered PCA method on the WRIM faults detection compared to some other signal analysis method.