Phase estimation of the start-up signal based on the fractional delay filter

Start-up signal analysis is an effective way to recognize the latent faults for rotating machinery. In this way, phase estimation is the key. Current methods cannot provide an accurate estimation for the phase due to the signal in start-up is non-stationary. A fractional delay filter based method is proposed to analyze the start-up signal for phase estimation. This method can implement the fractional interpolation of the discrete sample points. Therefore, it can convert the free sampling signal to the angle sampling signal which can estimate the phase accurately. The simulating signal and the experiment result are applied to validate the effectiveness and the accuracy of this method.