Probabilistic analysis and fatigue life assessment of floating collar of fish cage due to random wave loads

Abstract Fish farms have moved into the open sea in recent years and the failure risk of fish cages has increased in the harsher environmental conditions. Fatigue failure is an important limit state for the floating system of the fish cage under the long-term action of waves. A fatigue analysis of the high-density polyethylene (HDPE) floating system in random waves is presented to investigate the applicable probability density function for estimating the floating system fatigue life in this study. The stress response of the floating collars is firstly analyzed based on the finite element analysis combined with a hydrodynamic model. The stress histories of floating collars under each sea state are counted using the rainflow method as a benchmark for frequency domain fatigue analysis. The short- and long-term distributions of the stress ranges were fitted by various probability density functions including Rayleigh, Weibull, Gamma and generalized extreme value (GEV) distributions. Comparisons of the estimated fatigue life using different short- and long-term distributions were performed. Results indicate fatigue estimations based on the GEV and Gamma distributions of the short-term stress range by removing the negligible low stress range give much more accurate fatigue damage results than the traditional Rayleigh distribution. Weibull distribution overestimates the fatigue lifetime of the floating collar. The fatigue lifetime based on the long-term distribution was evaluated and was compared with the rainflow statistic and deterministic method results. The results show the long-term distribution is much more conservative for estimating the fatigue lifetime of the floating system.