Estimation of remaining useful life of bearings using sparse representation method

Prognostics and health management (PHM) play an important role in improving the reliability and safety of systems in industry. A method using sparse representation to solve the PHM questions raised in the IEEE 2012 PHM Data Challenge Competition is presented. It is discovered that the sparse representation score could be a good health indicator of ball bearings compared with traditional acceleration signal or some extracted features like energy, root mean square, maximum peak value and skewness. A group of experimental datasets from seventeen ball bearings are provided by the FEMTO-ST Institute for the competition. The data consists of six groups of bearing acceleration data for algorithm training and another eleven groups of bearing accelerated data for testing. The training data groups are used to build the full score cures model in order to estimate the remaining useful life of the truncated test data groups of the ball bearings. The result is compared with the result using a mixture of traditional features like energy, maximum peak value, root mean square and skewness. Result using the sparse representation method is relatively promising and could well reflect the states of the rotating ball bearings.