﻿ 浅水AUV耐压舱多目标优化设计
 舰船科学技术  2022, Vol. 44 Issue (5): 59-64    DOI: 10.3404/j.issn.1672-7649.2022.05.012 PDF

Multi-objective optimization design of shallow water AUV compressive cabin
WANG Yang-bin, WANG Li-jun, WANG Ru-xuan, XU Da, TIAN Bao-qiang
School of Mechanical Engineering, North China University of Water Resources and Electric Power, Zhengzhou 450000, china
Abstract: Aiming at the problems of conservative calculation and difficult experiments in the design process. The structure of the compressive cabin is designed. On the premise of meeting the requirements of structural strength and stability, the simplified model of the compressive cabin is optimized and analyzed based on the design exploration in Workbench. The optimal space-filling design (OSF) experiments tape is used to obtain the sample space of the optimization variables, and the neural network is used to construct the optimization response surface model; Based on the application of Multi-Objective genetic algorithm (MOGA) in the model, the optimal design size is obtained. The simulation results show that under the condition of good strength and stability, the mass and volume of the optimized compressive cabin are significantly reduced, and the lightweight design is realized.
Key words: compressive cabin     response surface     genetic algorithm     optimization design
0 引　言

1 耐压舱的设计与建模

 $t=\beta \frac{{P}_{cr}\cdot R}{{\sigma }_{s}} ，$ (1)
 ${L}_{cr}=\frac{2.59\times \left(1-{\vartheta }^{2}\right)\times D}{2}\times \sqrt{\frac{D}{t}} 。$ (2)

 图 1 AUV耐压舱结构图 Fig. 1 The figure of the AUV compressive cabin
2 静力学仿真求解

 图 2 静力学仿真云图 Fig. 2 The nephogram of Statics simulation

3 基于Workbench的耐压舱多目标优化

 图 3 多目标优化流程图 Fig. 3 Multi-objective optimization flow chart
3.1 耐压筒的多目标优化 3.1.1 优化数学模型与DOE设计试验

3.1.2 利用神经网络构建响应面模型

 图 4 响应面图示 Fig. 4 Response surface model

3.1.3 利用多目标遗传算法求优化解

 图 5 耐压筒优化后的变形、应力云图 Fig. 5 Cloud image of optimized pressure cylinder

3.2 半球封头和后盖的优化设计

3.3 优化结果

4 结　语

1）设计了一种半球壳与圆板球头，通过螺栓与双径密封器连接到圆柱主舱体的AUV耐压舱结构，通过理论计算，确定耐压舱初步的尺寸，建立参数化的三维模型。

2）利用Workbench中的Static structural模块，对基于理论计算的耐压舱结构进行仿真求解，得到总变形量、等效应变、等效应力分布云图，同时满足强度与稳定性要求，但是结果较为保守，可见后续的优化分析十分重要。

3）在强度和稳定性均满足要求的前提下，通过Workbench的Design Exploration模块对耐压舱各个部分进行优化设计，且通过响应面定性分析了变量之间的敏感性关系。利用多目标遗传算法（MOGA）得到优化候选结果，优化后的耐压舱在有足够强度和稳定性的前提下，质量和体积分别减小了30.6%,实现了轻量化设计，降低了生产成本，缩短了设计周期，优化结果弥补了理论计算的不足，验证了本文提出的优化设计方法和理论和合理性和正确性。

 [1] 谢鹏, 高志旺, 王志东, 等. 基于多种组合优化算法外形优化设计[J]. 舰船科学技术, 2021, 43(13): 45-48. [2] 高原, 黄进浩, 王永军, 等. 基于拓扑优化和尺寸优化的水下耐压球壳轻量化设计[J]. 舰船科学技术, 2019, 41(21): 54-58. [3] 甘盛霖, 梅益, 罗宁康, 等. 基于多目标的磨床床身结构优化设计[J]. 现代制造工程, 2021(3): 132-137. [4] 高启升, 朱兴华, 于延凯, 等. UUV耐压结构多目标优化设计[J]. 工程设计学报, 2020, 27(2): 232-238. [5] 董华超, 宋保维, 王鹏. 水下航行器壳体结构多目标优化设计研究[J]. 兵工学报, 2014, 35(3): 392-397. DOI:10.3969/j.issn.1000-1093.2014.03.015 [6] 黄维, 刘惟伊, 刘志恩, 等. 基于多目标遗传算法的实验目标车底盘结构优化[J]. 工程设计学报, 2021, 28(1): 80-88. [7] 张洪彬, 徐会希, 陈仲. 水下机器人耐压舱优化设计与结构分析[J]. 海洋学研究, 2018, 36(3): 84-88. DOI:10.3969/j.issn.1001-909X.2018.03.009 [8] YANG Zhao-qi, CAO Yong-hui, LIU Jing. A buckling analysis and optimization method for a variable stiffness cylindrical pressure shell of AUV[J]. Journal of Marine Science and Engineering, 2021, 9(6): 637-637. DOI:10.3390/jmse9060637 [9] HE Yan-ru, SONG Bao-wei, DONG Hua-chao. Multi-objective optimization design for the multi-bubble pressure cabin in BWB underwater glider[J]. International Journal of Naval Architecture and Ocean Engineering, 2018, 10(4): 439-449. DOI:10.1016/j.ijnaoe.2017.08.007 [10] 刘喆, 黄中华, 吴荣杰, 等. 基于遗传算法的耐压舱优化设计[J]. 制造业自动化, 2020, 42(12): 86-89. DOI:10.3969/j.issn.1009-0134.2020.12.017 [11] 何衍儒, 宋保维, 曹永辉. 翼身融合自主式水下航行器的多泡结构耐压舱分步优化设计[J]. 西北工业大学报, 2018, 36(4): 664-670. DOI:10.1051/jnwpu/20183640664 [12] 严高超, 沈孝芹, 于复生, 等. 基于ANSYS-Workbench对水下机器人框架结构优化设计[J]. 制造业自动化, 2020, 42(10): 1-3+7. DOI:10.3969/j.issn.1009-0134.2020.10.001