Automatic Identification of Ship-Ice Impact Based on SDD Model

When polar ships navigate icy areas, their hulls encounter sea ice collisions, generating significant structural vibrations. These vibrations pose considerable challenges to the overall safety of the ship and the stability of shipboard equipment. Different from local structural vibration, the global motion and vibration response of the ship hull can be collectively called shipice impact, which has a greater impact on the functionality of shipboard equipment. Based on the data measured by distributed accelerometers on board the Xuelong II during its voyage in the Antarctic ice area, this paper carries out the study of icebreaking shock, and utilizes the deep learning method to identify the ice-breaking vibration events. Through spectral analysis, the ship-ice impact signal is separated, and an ice-breaking impact identification and analysis method is established. In this paper, an improved single point target detection (SSD) based model is proposed to recognize ship-ice impact in vertical acceleration. The results show that the model can accurately recognize ship-ice shock in acceleration with an accuracy of 87.15%. The model offers valuable insights for selecting polar ship equipment, thereby advancing the development of intelligent polar ships.