Mechanical fault prediction based on principal component analysis

In order to find an effective way to predict running of mechanical equipment, a new prediction method of mechanical fault based on principal component analysis (PCA) is proposed in this study. The disadvantages of traditional prediction methods and the presence and development in mechanical fault of PCA-based predication method were briefly introduced. Theoretical basis, analysis processes and parameters selection of PCA were also investigated. It is proved that PCA-based prediction neglects the structure and principle of system in detail and only depends on the data from the sensor. Simulation experiments show that the PCA-based prediction method is a promising and feasible way to forecast mechanical failure, because the algorithm is simple and easy to implantation, and it can reduce noise and simplify data processing.