Title :
Condition monitoring based on kernel classifier ensembles
Author :
Mendel, Eduardo ; Varejão, Flávio M. ; Rauber, Thomas W. ; Batista, Rodrigo J.
Author_Institution :
Dept. of Comput. Sci., Univ. of Espirito Santo, Vitória, Brazil
Abstract :
The objective of this work is the model-free diagnosis of faults of motor pumps installed on oil rigs by sophisticated kernel classifier ensembles. Signal processing of vibrational patterns delivers the features. Different kernel-based classifiers are combined in ensembles to optimize accuracy and increase robustness. A comparative study of various classification paradigms, all performing implicit nonlinear pattern mapping by kernels is done. We employ support vector machines, kernel nearest neighbor, Bayesian Quadratic Gaussian classifiers with kernels, and linear machines with kernels.
Keywords :
Bayes methods; Gaussian processes; condition monitoring; electric motors; fault diagnosis; mechanical engineering computing; pattern classification; pumps; signal processing; support vector machines; vibrations; Bayesian quadratic Gaussian classifiers; condition monitoring; fault diagnosis; motor pumps; nonlinear pattern mapping; oil rigs; signal processing; sophisticated kernel classifier ensembles; support vector machines; vibrational patterns; Fault diagnosis; Feature extraction; Kernel; Mathematical model; Polynomials; Support vector machines; Vibrations;
Conference_Titel :
Industrial Informatics (INDIN), 2011 9th IEEE International Conference on
Conference_Location :
Caparica, Lisbon
Print_ISBN :
978-1-4577-0435-2
Electronic_ISBN :
978-1-4577-0433-8
DOI :
10.1109/INDIN.2011.6034841