This study aims to scientifically assess the disaster risks of ship net-type cages in wave environments, address the challenges in disaster prevention and control under extreme wave conditions, and promote the development of the offshore aquaculture industry. Numerical simulations were conducted using the AQWA hydrodynamic module in ANSYS to simulate the dynamic processes of ship-type net cages under various structural and wave conditions. After obtaining the data, a neural network algorithm was employed to construct the nonlinear relationship between disaster factors and structural damage, while the grey relational analysis method was used to identify the dominant disaster-causing factors. The results show that the structural motion responses and dynamic loads calculated by the numerical model closely match the test results, with an error of no more than 10%. The established neural network model accurately predicts the dynamic disaster situations, with a prediction error of no more than 5% and a root mean square error of no more than 0.52. It was determined that wave height is the dominant factor for mooring line breakage, and the floating frame length and wave height are the dominant factors for floating frame cracking. The research demonstrates that the neural network model can effectively predict the disaster damage for ship-type net cages and provides significant support for mooring line selection and floating frame safety assessment.
