A novel ball bearing fault diagnosis approach based on auto term window method

•Rolling element bearing is the most important components in the rotating machinery.•This paper introduces a new fault diagnosis method into the ball bearing diagnosis.•The new method could identify the fault accurately and increase the visibility of ball bearing failure signals.•The ATW method can also be used to analyze some other non-stationary vibrations of rotating machinery.

A new bearing fault diagnosis method based on auto term window (ATW) method is proposed in this paper. Ball bearing is the foremost important and also much easier to be damaged component in the rotation machinery. Vibration signature analysis of machine components is a commonly fault detect technique employed in ball bearing systems. The new fault diagnosis method proposed in this paper is applied to extract fault features in bearing vibration signals. On the base of the Wigner–Ville distribution (WVD) analysis of the bearing signal, the ATW method can not only suppress cross terms effectively, but also strengthen the energy of the auto terms and enhance the feature extraction effect, which is important in the following fault diagnosis research. The ball bearing fault experiment results proved the effectiveness of the ATW method in the last section of this paper.