analysis and interpretation of bearing vibration data using principal component analysis and self - organizing map

induction motor bearing is one of the key parts of the machine and its analysis and interpretation are important for fault detection. in the present work vibration signal has been taken for the classification i.e. bearing is healthy (h) or defective (d). for this purpose, clustering based classification of bearing vibration data has been carried out using principal component analysis (pca) and self organising map (som). from the vibration signal, twelve statistical features have been extracted from both the healthy and the defected condition of the bearing. further, these data are subjected to pca to extract significant features relevant to cluster structure. it is observed that out of twelve features only four features are found significant which is feed to the som model.  the som based classification is able to achieve an accuracy of 100%. this cluster-based method of feature reduction and classification could be useful in assessing the induction motor incipient bearing fault detection with large data set.