Exploring the distribution of time to failure of bearing experimental data

Abstract

Abstract Bearing failure degradation describes the probability of failure after survive a length of time. This paper explores the empirical distribution of time to failure and modeling of bearing failure degradation using vibrational experimental data. Kurtosis was used as a feature of vibrational data. The failure probability was estimated using actuarial method. Fitting to some time to failure distributions and bearing failure degradation models will be discussed as an illustration of the proposed method. The results of the study can be used to predict the percentage bearing failures for a certain time of period. The results of the proposed method can contribute to maintenance system of a mechanical system