Fault Detection of Bearings Using a Rule-based Classifier Ensemble and Genetic Algorithm

Abstract

This paper proposes a reduct construction method based on discernibility matrix simplification. The method works with genetic algorithm. To identify potential problems and prevent complete failure of bearings, a new method based on rule-based classifier ensemble is presented. Genetic algorithm is used for feature reduction. The generated rules of the reducts are used to build the candidate base classifiers. Then, several base classifiers are selected according to their diversity and the scale of them. Weights of the selected base classifiers are calculated based on a measure of support rate. The classifier ensemble is constructed by the base classifiers. The accuracy reached 98.44% which is 4.5% higher than that of the three base classifiers.