Reliability assessment for ultimate longitudinal strength of ship hulls in composite materials

A method for reliability assessment of the ultimate longitudinal strength of ship hulls in composite materials is described. The ultimate longitudinal strength of a ship hull is predicted by a progressive collapse analysis based on load-average strain curves of stiffened composite panels that are developed by a progressive failure nonlinear finite element analysis. Sagging, hogging and slamming conditions are regarded as three extreme loading conditions and the corresponding limit state functions are established with an appropriate stochastic modeling of the basic design variables, such as the modeling uncertainties, the materials properties, the lamina thickness, the lamination angle, the still-water moment, the wave-induced moment, and the slamming-induced moment. The reliability estimation is achieved by an improved first-order reliability algorithm. The sensitivities of the reliability estimates with respect to model parameters are determined and those parameters with small sensitivity factors are replaced by deterministic values to improve the computational efficiency in the prediction of the ultimate longitudinal strength of ship hull. An all-composite ship is analyzed for demonstration. (c) 2007 Elsevier Ltd. All rights reserved.