Optimization of manned submersibles based on the concurrent subspace design of response surface

Abstract

To improve the design efficiency and comprehensive performance of manned submersibles, the design parameters of one type of manned submersible were determined; the relationship between disciplines and systems was analyzed; the discipline decomposition of the manned submersible was completed on the basis of multidisciplinary design optimization; and the design structure matrix of manned submersibles was established. A response surface model was used to construct the discipline analysis models of ship type, propulsion discipline, structural discipline, and other disciplines, and a system-level analysis model was established by using a formula. By taking the weight of a submarine as the objective function, the optimization frameworks of the manned submersible based on response surface concurrent subspace design (CSD-RS) and the concurrent subspace optimization (CSSO-RS) were established. These frameworks could update the approximation model during optimization to improve the efficiency and accuracy of the solution. The two frameworks mentioned above were used to solve the optimization problem. The results showed that compared with the weight reduction of the initial scheme, that of CSD-RS was 7.822%,and that of CSD-RS was 7.311%. However, CSD-RS has more advantages in accuracy, efficiency, and probability of obtaining the optimal solution than the other methods.