Optimal design and strength reliability analysis of pressure shell with grid sandwich structure

Abstract

This study focuses on lightweight optimum design of head pressure shell of an autonomous underwater vehicle (AUV) with reliability requirements. The grid sandwich structure is applied to the pressure shell for lightweight. An optimal mathematical model is established, which takes weight and the maximum von-Mises stress as objective and strength requirement as a constraint condition. This optimization model is solved by genetic algorithm (GA) based on back propagation (BP) neural network surrogate model, which is optimized by particle swarm optimization (PSO) for improving accuracy. Additionally, the randomness of geometry sizes, material properties and loads are considered to ensure reliability. A stress-strength interference model is established for obtaining failure probability and reliability according to statistical regularity of yield strength and the maximum von-Mises stress. The result shows that the weight of grid sandwich structure pressure shell is reduced by 38.26% compared with the solid pressure shell. The width of transverse grid has more influence on the maximum von-Mises stress than longitudinal grid. The critical load factor is 21.611 and buckling deformation will not occur under given conditions. The pressure shell with grid sandwich structure meets three sigma level with random variables, which can ensure enough reliability of AUV. The above analysis shows that application of grid sandwich structure to the pressure shell of AUV head is feasible.