Riser Configuration Design for Extreme Shallow Water With Surrogate Model Based Optimization

The aim of this paper is to study the optimization design of a steep wave riser for extreme shallow water based on radial basis function (RBF) surrogate model approach. As the design of riser configuration is rather time consuming and exhaustive due to the nonlinear time domain analysis and large quantities of load cases, it would be more difficult when we need to deal with some extreme design such as in extreme shallow water. The surrogate model in this paper is constructed with RBF networks from the samples obtained by optimal Latin hyper cubic sampling and time domain analysis in a given design space. Then, a hybrid optimization is performed based on the established surrogate model. An optimized design is finally found to meet the design criterion with high accuracy and efficiency, even all the samples fail to meet the curvature criterion.