INTELLIGENT SYSTEM FOR CONDITION CLASSIFICATION OF ROLLING ELEMENT BEARINGS Martina Joševska In this work the intelligent system for condition classification of rolling element bearings is presented. The system is based on the application of two-class Support Vector Machine classifier. Each observed bearing needs to be classified either as damaged or as healthy. Before designing and testing the classifier the preparation of input data has to be performed. Input data for the binary SVM classifier are obtained by recording vibration signals from 196 rolling element bearings. Based on the measured data the statistical and frequency properties are calculated. These properties are called features. Each bearing is characterized with 18-dimensional features vector. However, due to the high dimensionality of the features vector the number of features has to be reduced so that high-quality classification can be achieved. In order to reduce the initial number of features three linear techniques for feature extraction have been implemented: Linear Discriminant Analysis- LDA, Principal Component Analysis - PCA, Independent Component Analysis- ICA and the analysis of classification results was carried out.
Two OPTEC professors have been awarded three "Gouden Krijtjes", the yearly teaching awards given by the organization of engineering students (vtk). Prof. Lombaert was awarded the prize for the best course in civil engineering, and Prof. Diehl the prizes for the best professor and the best course in mathematical engineering (where he teaches numerical optimization). They received these awards at the yearly "proffentap" where experienced students taught them how to draft beer professionally.
