An Approach of Multidisciplinary Optimization to Underwater Vehicle Profile Design

Abstract

The parallel multidisciplinary optimization design is proposed in this paper to improve the hydrodynamic performance of underwater vehicle. An isight optimization platform is constructed in which the optimal Latin hypercube algorithm is selected as the experimental design method, and the RBF neural network is used to construct the approximate model. The simulated annealing algorithm is taken as the optimization algorithm. By using the optimization strategy proposed, the drag and energy consumption are significantly reduced, which demonstrates the validity of the proposed optimization method.