Diagnosis of the Industrial Systems by Fuzzy Classification

Abstract

The aim of this paper is to present a classifier based on a fuzzy inference system, the final decision being made with the help of a thresholder. For this classifier, a parameterization method which is not necessarily based on an iterative training is presented. This approach can be seen as a pre-parameterization of the classifier which allows the building up of a base set of rules, and initialise the parameters of membership function. A continuous and derivable version of the previous classifier is also presented. For this last classifier an iterative learning algorithm based on a gradient method is suggested. An example using the learning basis IRIS, which is a benchmark for the problems of classification, is presented. This example allows us to compare the performance of this classifier with the works of other authors. Finally this classifier is applied to the diagnosis of a D.C. motor showing the utility of this method.