水下无人系统学报2017，Vol.25Issue(6)：410-417,8.DOI:10.11993/j.issn.2096-3920.2017.05.003
使用混合代理模型的自主式水下航行器藕节壳体多目标优化
Mixture Surrogate Model Based Structural Optimization Design of Multiple Intersecting Spheres for Automatic Undersea Vehicle
摘要
Abstract
To solve the problem that the existing structural optimization designs of multiple intersecting spheres are simplified models and are lack of internal effective volume analysis,this paper uses the non-dominated sorting genetic algorithm Ⅱ(NSGA-Ⅱ) and the mixture surrogate model constructed by radial basis function,support vector regression function and Kriging function to perform multi-objective optimization of the tangent arc connective and ring-stiffened multiple intersecting spheres of an autonomous undersea vehicle(AUV) in terms of the buoyancy-weight ratio and the internal effective volume of the spheres.Parametric modeling of the two multiple intersecting spheres are implemented using UG secondary development.The Pareto front of this multi-objective optimization problem is obtained under the constraint of satisfying maximum equivalent stress and bulking factor.This research may provide a reference for optimization design of AUV shell.关键词
自主式水下航行器(AUV)/藕节壳体/多目标优化/混合代理模型/NSGA-ⅡKey words
automatic undersea vehicle(AUV)/multiple intersecting spheres/multi-objective optimization/mixture surrogate model/non-dominated sorting genetic algorithm Ⅱ(NSGA-Ⅱ)分类
交通工程引用本文复制引用
何衍儒,宋保维,曹永辉..使用混合代理模型的自主式水下航行器藕节壳体多目标优化[J].水下无人系统学报,2017,25(6):410-417,8.基金项目
国家自然科学基金资助项目(51375389). （51375389）
