Most of the currently available methods for fault detection in rotational machines assume constant and known operating conditions. In the underlying paper we present a method based on wavelet transform that allows effective fault detection under variable operating conditions. The method is based on the computation of two statistical features obtained from the wavelet coefficients. With such an approach we were able to detect various gear and bearing faults on a test-bed with two-stage gearbox.
COBISS.SI-ID: 24084007
The dynamics of damage propagation can be modeled as a nonlinear stochastic dynamical state-space process, where system states are not directly accessible and all information about them has to be inferred from the output data. The paper employs an iterative procedure for calculating the Maximum-Likelihood estimate of model parameters, based on the Expectation- Maximization algorithm together with the unscented Kalman filter for estimation of system states. The algorithm has been used to predict the remaining useful life of a single-stage gearbox system.
COBISS.SI-ID: 23769639