Residual fatigue life prediction for ship propeller based on test signal characteristic fusion and TV-HSMM: An experimental case study

It is a key issue to investigate the degradation law of ship propeller service performance and predict its residual fatigue in marine engineering. In this paper, the underwater test are carried out and fracture characteristics are detected to systematically analyze the failure mechanism of the ship propeller in operating states. Then, a feature fusion method based on adaptive popular learning algorithm is proposed, the redundant information in multidomain characteristics set is eliminated, and the fusion characteristic index of propeller performance degradation trend is constructed according to the LLE method. Finally, based on the TV-HSMM method, the state transformation coefficient is optimized, and the propeller operating state recognition and residual fatigue life prediction are realized with the combined application of the feature fusion data. The research on propeller residual fatigue life prediction based on data fusion is of great significance to avoid major faults and ensure the operation safety of ships.