首页 | 本学科首页   官方微博 | 高级检索  
     

基于克里金模型的潜水器耐压舱结构优化
引用本文:张磊,胡震. 基于克里金模型的潜水器耐压舱结构优化[J]. 船舶力学, 2020, 0(1): 108-117
作者姓名:张磊  胡震
作者单位:中国船舶科学研究中心水下工程研究室
基金项目:国家重点研发计划项目“全海深载人潜水器总体设计、集成与海试”(2016YFC300600)
摘    要:本文研究Kriging建模和序贯加点相结合的方法,用于解决目标函数或者约束函数无法显式表达下的优化问题。在加点准则中为了避免出现龙格现象,改进了加点优化时步长要求,在优化过程中融入多种群遗传算法提高了序贯加点值的计算准确性。计算结果表明,改进后的加点准则在存在多区域或者多个极值点的情况下均可快速搜索到最优点位置。建立了载人潜水器载人舱以重量为优化目标,以强度要求为约束的球壳厚度优化模型,采用此方法在初始选取5个样本取样,序贯加点6个样本点,计算得到球壳的最优厚度值为50.5569 mm,可有效指导设计工作。

关 键 词:载人潜水器  多种群遗传算法  厚度优化  约束函数  极值点  强度要求  球壳  计算准确性

Structural optimization of submersible spherical pressure hull based on Kriging model
ZHANG Lei,HU Zhen. Structural optimization of submersible spherical pressure hull based on Kriging model[J]. Journal of Ship Mechanics, 2020, 0(1): 108-117
Authors:ZHANG Lei  HU Zhen
Affiliation:(China Ship Scientific Research Center,Wuxi 214082,China)
Abstract:This paper studies the combination of Kriging modeling and sequential addition points,which is used to solve the optimization problem that the objective function or constraint function cannot be expressed explicitly.In order to avoid the phenomenon of the dragon lattice in the plus point criterion,the step size requirement of the point optimization is improved,and the multi-group genetic algorithm is integrated into the optimization process to improve the calculation accuracy of the sequential addition point value.The calculation results show that the improved add-on criterion can quickly search for the most advantageous position in the presence of multiple regions or multiple extreme points.The spherical shell thickness optimization model with the weight as the optimization target and the strength requirement as the constraint is established.Five sample samples are selected and six sample points are added in sequence to calculate the spherical shell.The optimum thickness value is 50.5569 mm,which can effectively guide the design of the submersible pressure hull.
Keywords:Kriging model  agent model  submersible  structural optimization
本文献已被 CNKI 维普 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号