Accidents involving chemical tankers are often highly hazardous. The operational status of cargo tanks and loading/unloading systems is critical to the safety of these ships during service. This paper reviews 62 accident cases related to the cargo tanks and loading/unloading systems of chemical tankers occurring between 2003 and 2021. A statistical analysis of the accident causes is conducted from four perspectives: accident type, deadweight tonnage of the involved ship, ship age, and the ship's operational status at the time of the accident. This analysis aims to summarize the conditions under which these accidents occur. Thirty-one typical cases are selected for a more detailed examination. The potential risks leading to these accidents are categorized into human factors and non-human factors. Finally, targeted control measures are proposed, including suggestions for chemical tanker rules and standards, as well as recommendations for the management and operational requirements of shipping companies. These measures are intended to provide technical support for enhancing the safety of chemical tanker transportation.

This paper focuses on the acoustic fault diagnosis of the ship centrifugal pump system based on the vibration threshold in view of the insufficient connection between the acoustic fault diagnosis threshold and the historical measured data during the ship construction stage. Firstly, the analysis of the test data of the full-scale ship shows that the maximum vibration fluctuations of the ship centrifugal pump system is more than 10 dB due to the differences in boundary conditions and equipment. Secondly, a normal distribution assumption is made for the vibration deviation value, which is verified by the statistical analysis of the test data. Finally, a threshold extraction method is proposed based on the normal distribution interval estimation principle to check the validity of the above thresholds by simulation experiments. The results show that the diagnostic accuracy of the threshold extracted by this method is 99.45% for normal operating conditions and 92.3% for fault operating conditions. This study can provide a reasonable and operable acoustic fault diagnosis method with high accuracy for the centrifugal pump system during the ship construction stage.