基于蠕变特性的耐压球壳多目标优化  

Optimization design for the pressure spherical shell of manned submersible with consideration of creep of window

作　　者：黄浔 韩端锋[1] 赵方瑜 刘峰[1] HUANG Xun;HAN Duan-feng;ZHAO Fang-yu;LIU Feng(College of Shipbuilding Engineering,Harbin Engineering University,Harbin 150001,China;College of Nuclear Science and Technology,Harbin Engineering University,Harbin 150001,China)

机构地区：[1]哈尔滨工程大学船舶工程学院,哈尔滨150001 [2]哈尔滨工程大学核科学与技术学院,哈尔滨150001

出　　处：《船舶力学》2020年第3期362-370,共9页Journal of Ship Mechanics

基　　金：国家自然科学基金资助项目(51709063);国家科技重大专项资助项目(2011ZX05027-005);中央高校基本业务费资助项目(HEUCFP201603,HEUCFP201735,HEUCFM180106)。

摘　　要：本文针对载人潜器耐压球壳,在考虑了观察窗蠕变效应的基础上,设计了参数化计算流程,进行了设计变量的灵敏度分析,采用径向基神经网络建立了近似模型。进一步以临界应力、临界压力为约束,以球壳质量、主侧观察窗蠕变位移为目标,建立了多目标优化模型,采用第二代非支配排序遗传算法(NSGA-II)进行了求解。结果表明:在耐压壳体质量增加幅度不大的情况下,观察窗蠕变变形优化效果明显,大大提高了载人潜水器的安全性,为载人潜水器耐压球壳设计提供了参考。Aimed at the pressure spherical shell of manned submersible,and with creep effect of the window taken into consideration,a process of calculation based on parameterized models was designed.An approximate model was built by using radial basis functions network model.Furthermore,the angle and thickness of window were chosen as design variables,constrained by critical stress and pressure,the weight of the pressure spherical shell and the creep displacement of the observation window were taken as the target,the multiobjective optimization model was established and the multi-objective opimization problem was solved by the fast and elitist non-dominated sorting genetic algorithm(NSGA-Ⅱ).Results show that when the quality of pressure spherical shell increased modestly,the optimization effect of creep deformation is remarkable,the safety of manned submersibles is improved significantly.The research conclusion of this paper provides reference for the design of pressure hull of manned submersible.

关 键 词：载人潜器 观察窗 蠕变 耐压球壳 多目标优化 

分 类 号：U661.4[交通运输工程—船舶及航道工程]