Ship propulsion shafting bearing fault diagnosis based on holographic SDP similarity visual recognition

  Objectives  To address the problem of the fault diagnosis of ship propulsion shafting bearings, this paper proposes a visual diagnosis method based on a holographic symmetrical dot pattern (SDP) and similarity recognition.

  Methods  First, the bearing vibration signals are collected in three directions to comprehensively monitor the non-stationary changes in the time-domain and frequency-domain caused by the regular impact of the bearing faults. Second, based on SDP, multiple one-dimensional time-domain signals and spectrums are merged into a two-dimensional image to amplify the difference between different state signals. Finally, a simple bearing diagnosis is performed based on the similarity recognition method.

  Results  The results of engineering experiments show that this method can achieve the effective graphic fusion of multiple signals, fully display the characteristics of equipment signals and accurately diagnose faults.

  Conclusions  The results of this study can provide valuable references for the visual fault diagnosis of ship propulsion shafting bearings.