本文以超大潜深的潜艇耐压壳结构为研究对象，利用排水量表达式估算潜艇主尺度。通过Ansys软件的Apdl语言建立环肋锥柱壳的有限元模型，并分析计算耐压壳的强度及稳定性。以肋骨间距、耐压壳厚度和肋骨尺寸作为离散设计变量，以结构重量、总体失稳临界压力作为优化目标，实现基于神经网络和遗传算法的环肋锥柱壳多目标优化设计。在Matlab平台上，首先用拉丁超立方体抽样，再用BP神经网络建立起样本点和目标函数之间的映射关系，构建神经网络代理模型，最后调用多目标优化函数gamultiobj进行优化。优化结果表明，利用BP神经网络和遗传算法相结合进行复杂模型环肋锥柱壳的多目标优化，效率较高，精度较好，达到较理想的优化效果。
In this paper, a ultra-deep submarine pressure shell is taken as the research object, and the main scales of the submarine are estimated by using the displacement expression. The finite element model of the ring stiffened cone shell was established by the APDL language of Ansys software, and the strength and stability of the pressure shell were analyzed. The stiffener spacing, the pressure shell thickness and the stiffener size are taken as discrete design variables. The structural weight and the overall instability critical pressure are used as the optimization targets to realize the multi-objective optimization design of the ring stiffened cone shell based on neural network and genetic algorithm. Using the Matlab, Latin hypercube is firstly sampled, and then the BP neural network is used to establish the mapping relationship between the sample points and the objective function, and the neural network proxy model is constructed. Finally, the multi-objective optimization function gamultiobj is called to optimize. The optimization results show that BP neural network cooperated with genetic algorithm in solving the multi-objective optimization of complex ring stiffened cone shells are good in efficiency and precision.
2021,43(2): 6-12 收稿日期：2019-10-15
DOI：10.3404/j.issn.1672-7649.2021.02.002
分类号：U674.76
作者简介：李艳萍(1994-),女,硕士研究生,研究方向为船舶结构优化
参考文献：
[1] 潘涛. 深潜器耐压结构强度分析与优化设计[D]. 哈尔滨: 哈尔滨工程大学, 2010.
PAN Tao. Strength analysis and optimum design of compressive structure of deep submersible[D]. Harbin: Harbin Engineering University, 2010.
[2] 王燕. 潜艇结构的优化设计方法研究[D]. 哈尔滨: 哈尔滨工程大学, 2010.
WANG Yan. Research on the optimization design method of submarine structure[D]. Harbin: Harbin Engineering University, 2010.
[3] 刘栋. 改进微粒群算法在多目标优化问题中的应用[D]. 济南: 山东师范大学, 2008.
LIU Dong. Application of improved particle swarm optimization in multi-objective optimization problems. Application of improved particle swarm optimization in multi-objective optimization problems[D]. Jinan: Shandong Normal University, 2008.
[4] 李学斌, 朱学康. 潜艇耐压圆柱壳的多目标优化设计[J]. 中国造船, 2009, 50(1): 10-17
LI Xue-bin, ZHU Xue-kang. Multi-objective optimization design of submersible pressure-resistant cylindrical shell[J]. China Shipbuilding, 2009, 50(1): 10-17
[5] Guangyong SUN, Guangyao LI, Zhihui GONG, etc. Multiobjective robust optimization method for drawbead design in sheet metal forming[J]. Materials & Design, 2010, 31(4): 1917-1929
[6] 张彩坤, 陈国琳, 王磊. 国外核潜艇大潜深技术发展趋势[J]. 舰船科学技术, 2011, 33(12): 134-137
ZHANG Cai-kun, CHEN Guo-lin, WANG Lei. Development trend of large submersible technology for foreign nuclear submarines[J]. Ship Science and Technology, 2011, 33(12): 134-137
[7] 孟宪斌, 易彩虹, 吴小玲, 等. 钛及钛合金复合材料发展及工业应用[J]. 中国化工装备, 2013, 6: 3-7
MENG Xian-bin, YI Cai-hong, WU Xiao-ling, et al. Development and industrial application of titanium and titanium alloy composite materials[J]. China Chemical Equipment, 2013, 6: 3-7
[8] 程远胜, 孙莹, 闫国强, 等. 基于神经网络与遗传算法的潜艇舱壁结构优化[J]. 中国造船, 2008, 49(4): 81-86
CHENG Yuan-sheng, SUN Ying, YAN Guo-qiang, et al. Optimization of submarine bulkhead structure based on neural network and genetic algorithm[J]. China Shipbuilding, 2008, 49(4): 81-86
[9] 变色脸. 中国潜艇赶超世界先进, 完美静音性能让美国直呼不可能[EB/OL]. https://bbs.tiexue.net/post2_12673364_11.html, 2017-8-3.
Color changing face. Chinese submarine catches up with the world's advanced, perfect mute performance makes it impossible for the United States to call directly [EB/OL]. https://bbs.tiexue.net/post2_12673364_11.html, 2017-8-3.
[10] 陈明高, 石仲堃. 常规潜艇排水量和主尺度的确定新方法[J]. 中国舰船研究, 2006, 1(2): 38-41
CHEN Ming-gao, SHI Zhong-kui. A new method for determining the displacement and principal dimensions of conventional submarines[J]. China Ship Research, 2006, 1(2): 38-41
[11] GJB 64.2A-1997, 舰船船体规范 潜艇[S]. 北京: 国防科学技术工业委员会, 1997.
GJB 64.2A-1997, Ship hull specifications submarine[S]. Beijing: National Commission for Science, Technology and Industry for National Defense, 1997.
[12] GJB-Z 21A-2001, 潜艇结构设计计算方法[S]. 北京: 国防科学技术工业委员会, 1991.
GJB-Z 21A-2001, Calculation method of submarine structure design[S]. Beijing: National Commission for Science, Technology and Industry for National Defense, 1991.
[13] 陈皓. 遗传算法求解一类带工艺约束的并行机调度问题[D]. 武汉: 华中科技大学, 2005.
CHEN Hao. Genetic algorithm for a class of parallel machine scheduling problems with process constraints[D]. Wuhan: Huazhong University of Science and Technology, 2005.
[14] 朱小梅, 郭志钢, 杨先凤. 基于遗传算法BP神经网络优化证券组合投资[J]. 江汉大学学报(自然科学版), 2005, 33(3): 47-50
ZHU Xiao-mei, GUO Zhi-gang, YANG Xian-feng. Optimizing portfolio investment based on genetic algorithm BP neural network[J]. Journal of Jianghan University (Natural Science), 2005, 33(3): 47-50
[15] 刘猛. 云计算平台下神经网络方法研究[D]. 成都: 电子科技大学, 2011.
LIU Meng. Research on neural network method under cloud computing platform[D]. Chengdu: University of Electronic Science and Technology of China, 2011.
[16] 刘天舒. BP神经网络的改进研究及应用[D]. 哈尔滨: 东北农业大学, 2011.
LIU Tianshu. Improved research and application of BP neural network[D]. Harbin: Northeast Agricultural University, 2011.
[17] 卢纯, 石秉学. 采用BP-GA算法的一种LSI神经网络的电路设计[J]. 清华大学学报(自然科学版), 2001, 41(1): 103-106
LU Chun, SHI Bingxue. Circuit design of an LSI neural network using BP-GA algorithm[J]. Journal of Tsinghua University (Science and Technology), 2001, 41(1): 103-106
[18] 张宇, 黄小平, 闫小顺. 基于神经网络和粒子群算法的环肋圆柱壳优化设计[J]. 舰船科学技术, 2016: 38
ZHANG Yu, HUANG Xiao-ping, YAN Xiao-shun. Optimization design of ring-ribbed cylindrical shell based on neural network and particle swarm algorithm[J]. Ship Science and Technology, 2016: 38
[19] 陈果. 神经网络模型的预测精度影响因素分析及其优化[J]. 模式识别与人工智能, 2005, 18(5): 528-534. (3): 5-9.
CHEN Guo. Analysis and optimization of influencing factors on prediction accuracy of neural network models[J]. Pattern Recognition and Artificial Intelligence, 2005, 18 (5): 528-534. (3): 5-9.
[20] 马运义, 许建. 现代潜艇设计理论与技术[M]. 哈尔滨：哈尔滨工程大学出版社, 2012: 70.
