vibration signals in accelerated motor aging applicable For age determination WłaściW ości sygnałó W Wibracyjnych zWiązane z Widmem Fouriera W przyspieszonych badaniach starzenia się silnika mających na celu określenie jego Wieku
A series of computations based on the Fourier spectrum of vibration signals collected in artificial aging processes have been applied in order to tackle trends existing in the signals' sequence. In this study, features extracted from the (Fourier) spectrum of the vibration signal and expressing monotonic changes with respect to the motor age namely mean frequency, low order statistics of the power spectral density and cepstrum are used. Independence of the time domain data allows independent analysis. Results obtained in case of artificial aging are compared with results in case of controlled bearing faults with increasing severity. Clear distinctions are made and conclusions are drawn from the different behavioural patterns. Future research directions are indicated both in practical and theoretical sense.