Evaluación comparativa de métodos de clasificación empleados en el diagnóstico de fallos de procesos industriales (Comparative evaluation of classification methods used in fault diagnosis of industrial processes)

Alberto Prieto Moreno (albprieto@electrica.cujae.edu.cu)1, Orestes Llanes Santiago (orestes@electrica.cujae.edu.cu)1, José Manuel Bernal de Lázaro (jbernal@crea.cujae.edu.cu)1, Emilio García Moreno (egarciam@isa.upv.es)2

1Instituto Superior Politécnico José Antonio Echeverría
2Universidad Piltécnica de Valencia

This paper appears in: Revista IEEE América Latina

Publication Date: March 2013
Volume: 11,   Issue: 2 
ISSN: 1548-0992

This article presents a comparative study of the performance of classification techniques used for fault diagnosis in industrial processes. The techniques studied ranging from classifiers based on Bayes theory as Maximum a Posteriori Probability (MAP) and Nearest Neighbor (kNN) classifiers, through minimizing an objective function such as Artificial Neural Networks (ANN) and Support Machines Vector (SVM) and ending with the parameter estimation technique Partial Least Squares (PLS). Comparison of these techniques is based on the capacity of classification of the historical data and the generalization of new observations. Also, a discussion about the robustness of the classifiers against the dimensionality reduction process is presented. The study was conducted using the data from the testing process "Tennessee Eastman Process" (TEP).

Index Terms:
industrial processes, fault diagnosis, support vector machines, artificial neural networks, partial least squares, nearest neighbors classifier, MAP classifier   

Documents that cite this document
This function is not implemented yet.

[PDF Full-Text (325)]