水下发射水动力的多尺度预测网络研究

岳杰顺; 权晓波; 叶舒然; 王静竹; 王一伟

doi:10.6052/0459-1879-20-186

水下发射水动力的多尺度预测网络研究

doi: 10.6052/0459-1879-20-186

岳杰顺^*,
权晓波^†,
叶舒然^*,
王静竹^*,
王一伟^*,2), ,

^*中国科学院力学研究所流固耦合系统力学重点实验室,北京 100190
^†北京宇航系统工程研究所,北京 100076

基金项目: 1) 国家自然科学基金资助项目(11772340);国家自然科学基金资助项目(11672315)

详细信息

作者简介:
2) 王一伟,研究员,主要研究方向：高速水动力学、人工智能在流体力学中的应用. E-mail: wangyw@imech.ac.cn

通讯作者:
王一伟

中图分类号: TJ6,O352
计量
- 文章访问数: 1769
- HTML全文浏览量: 317
- PDF下载量: 594
- 被引次数: 0
出版历程
- 收稿日期: 2020-06-03
- 刊出日期: 2021-02-10

A MULTI-SCALE NETWORK FOR THE PREDICTION OF HYDRODYNAMICS IN UNDERWATER LAUNCH

^*Key Laboratory Mechanics in Fluid Solid Coupling Systems, Institute of Mechanics,Chinese Academy of Sciences,Beijing 100190,China
^†Beijing Institute of Astronautical Systems Engineering,Beijing 100076,China

摘要

摘要: 空泡的演化和水动力特征的预测在航行体发射的设计中有非常重要的意义. 人工智能技术已经成为了参数预测的重要手段.为了能够快速预测航行体水下发射过程的尾部压力的复杂变化, 提出了一种多尺度深度学习网络.该网络模型以一维卷积网络(1DCNN)为基础,构建了一种编码-解码型网络结构,通过不同的采样频率将原始数据分解为光滑部分和脉动部分,进而训练低保真度的大尺度网络和高保真度的小尺度网络.从而实现对不同物理过程的响应和捕捉.首先,通过数值模拟获得了不同发射条件下的尾部压力曲线,并结合空泡的理论机理构建了具有物理性的输入数据集.其次,将数据集进行分解处理,分别训练了两个尺度的深度学习网络. 最终将两组输出数据整合在一起,建立了底部压力预测模型.并通过测试和验证说明本文提出的多尺度网络对于多种常见的发射条件,能够实现航行体受力特征的快速准确的预测,光滑曲线、压力突变、震荡的频率和幅值都和数值模拟的结果吻合.证明本文的方法能够为运动和弹道的预测提供依据.
- 水下发射 /
- 深度学习 /
- 多尺度 /
- 一维卷积 /
- 尾部压力
Abstract: The prediction of cavitation and the hydrodynamic characteristics plays a significant role in the design of the underwater launched vehicle. In recent years, the artificial intelligence technology has become an important prediction method for these parameters. In order to quickly predict the dramatic changes of the bottom pressure in the underwater launching process, a multi-scale deep learning network is developed. This neural network model is based on a one-dimensional convolutional network (1DCNN) and established with an encoding-decoding network structure. The input data set is decomposed into a smooth part and fluctuating part through different sampling frequencies. A large-scale low-fidelity network and a small-scale high-fidelity network are trained separately to achieve the response and capture of different physical processes. Firstly, the bottom pressure under different launch conditions are obtained through numerical simulation, and the mechanism of bubble dynamics is constructed as a physical input data. Secondly, the data set is decomposed into two parts to train deep learning networks with two different scales respectively. Finally, two sets of output data based on two networks are integrated to establish a full prediction model. Testing and verification indicate that this newly developed multi-scale network can realize the fast and accurate prediction of the hydrodynamics of the underwater vehicle under various usual launch condition. The predicted bottom pressure curve during any stage, including the smooth stage, the transitional stage, as well as the frequency and magnitude of oscillation are consistent with the numerical simulation results. As a result, this method can provide a basis for the prediction of motion and trajectory of the underwater vehicle.
- underwater launch /
- deep learning /
- multi-scale /
- one-dimensional convolutional /
- bottom pressure

HTML全文

参考文献(33)

[1]	王一伟, 黄晨光. 高速航行体水下发射水动力学研究进展. 力学进展, 2018,48(1):259-298 (Wang Yiwei, Huang Chenguang. Research progress on hydrodynamics of high speed vehicles in the underwater launching process. Advances in Mechanics. 2018,48(1):259-298 (in Chinese))
[2]	杜特专, 王一伟, 黄晨光等. 航行体水下发射流固耦合效应分析. 力学学报, 2017,49(4):782-792 (Du Tezhuan, Wang Yiwei, Huang Chenguang, et al. Study on Coupling Effects of Underwater Launched Vehicle. Chinese Journal of Theoretical and Applied Mechanics. 2017,49:782-792 (in Chinese))
[3]	李帅, 张阿漫, 韩蕊. 水中高压脉动气泡水射流形成机理及载荷特性研究. 力学学报, 2019,51(6):1666-1681 (Li Shuai, Zhang Aman, Han Rui. The mechanism of jetting behaviors of an oscillating bubble. Chinese Journal of Theoretical and Applied Mechanics. 2019,51(6):1666-1681 (in Chinese))
[4]	燕国军, 梁欣欣, 张健等. 水下航行体垂直发射环境流场与弹道耦合数值模拟研究. 节能技术, 2019,37(4):307-312 (Yan Guojun, Liang Xinxin, Zhang Jian, et al. Investigation on the vertical launching process of the underwater vehicle by couple the flow field and the hydro–ballistics. Energy Conservation Technology. 2019,37(4):307-312 (in Chinese))
[5]	王晓辉, 张珂, 褚学森等. 水下点火推进尾空泡振荡的研究. 船舶力学, 2020,24(2):136-144 (Wang Xiaohui, Zhang Ke, Chu Xuesen, et al. Research on the pressure oscillation process of tail bubble of underwater igniting propulsion. Journal of Ship Mechanics. 2020,24(2):136-144 (in Chinese))
[6]	李杰, 鲁传敬. 潜射导弹尾部燃气后效建模及数值模拟. 弹道学报, 2009(4):10-12 (Li Jie, Lu Chuanjing. The model of combustion gas bubble of submarine-launched missile and numerical simulation. Journal of Ballistics, 2009(4):10-12 (in Chinese))
[7]	张佳悦, 李达钦, 吴钦等. 航行体回收垂直入水空泡流场及水动力特性研究. 力学学报, 2019,51(3):803-812 (Zhang Jiayue, Li Daqin, Wu Qin, et al. Numerical investigation on cavity structures and hydrodynamics of the vehicle during vertical water-entry. Chinese Journal of Theoretical and Applied Mechanics. 2019,51(3):803-812 (in Chinese))
[8]	孙健. 水下垂向运动超空泡航行体弹道仿真研究. [硕士论文]. 哈尔滨:哈尔滨工业大学, 2013 (Sun Jian. Research on vehicle trajectory simulation of underwater vertical movement super cavitation vehicle. [Master Thesis]. Harbin: Harbin Institute of Technology, 2013 (in Chinese))
[9]	宋武超, 魏英杰, 路丽睿等. 基于势流理论的回转体并联入水双空泡演化动力学研究. 物理学报, 2018,67(22):315-330 (Song Wuchao, Wei Yingjie, Lu Lirui, et al. Dynamic characteristics of parallel water-entry cavity based on potential flow theory. Acta Physica Sinica. 2018,67(22):315-330(in Chinese))
[10]	田再克, 杨锁昌, 冯德龙等. 基于摄动理论的落点预测算法研究. 现代防御技术, 2014,42(3):86-90 (Tian Zaike, Yang Suochang, Feng Delong, et al. Impact point prediction algorithm based on perturbation theory. Modern Defense Technology. 2014,42(3):86-90 (in Chinese))
[11]	魏倩, 蔡远利. 一种基于神经网络的中制导改进算法. 西安交通大学学报, 2016,50(7):125-130 (Wei Qian, Cai Yuanli. A modified algorithm on the midcourse guidance based on bp neural network. Journal of Xi'an Jiaotong University, 2016,50(7):125-130 (in Chinese))
[12]	邵雷, 雷虎民, 张大元. 基于网格划分与 BP 网络的中制导弹道在线生成方法. 弹道学报, 2019,31(3):1-6 (Shao Lei, Lei Humin, Zhang DY. Trajectory generation for midcourse guidance based on grid partition and bp neural network. Journal of Ballistics. 2019,31(3):1-6 (in Chinese))
[13]	吴朝峰, 杨臻, 曹文辉等. 基于GA-BP 算法的外弹道落点误差预测. 兵器装备工程学报, 2019,40(12):67-71 (Wu Chaofeng, Yang Zhen, Cao Wenhui, et al. Prediction of external ballistic landing error based on GA-BP algorithm. Journal of Ordnance Equipment Engineering. 2019,40(12):67-71 (in Chinese))
[14]	张秦浩, 敖百强, 张秦雪. Q-learning 强化学习制导律. 系统工程与电子技术, 2020,42(2):414-419 (Zhang Qinhao, Ao Baiqiang, Zhang Qinxue. Reinforcement learning guidance law of Q-learning. Systems Engineering and Electronic, 2019,40(12):67-71(in Chinese))
[15]	宫兆新, 曹嘉怡, 鲁传敬等. 发射参数对航行体水下运动的影响研究//第十三届全国水动力学学术会议暨第二十六届全国水动力学研讨会文集. 青岛市, 2014-08-22 (Gong Zhaoxin, Cao Jiayi, Lu Chuanjing, et al. Effects of the vertical launching parameters on the underwater vehicle movement//The 13th National Hydrodynamics Conference and the 26th National Hydrodynamics Seminar. Qingdao, 2014-08-22 (in Chinese))
[16]	Kurtz J. Deep learning in fluid dynamics. Journal of Fluid Mechanics, 2017,814:1-4
[17]	Tang M, Liu Y, Durlofsky LJ. A deep-learning-based surrogate model for data assimilation in dynamic subsurface flow problems. Journal of Computational Physics, 2020,413(15):1-57
[18]	Jagtap AD, Karniadakis GE. Adaptive activation functions accelerate convergence in deep and physics-informed neural networks. Journal of Computational Physics, 2020,404:1-23
[19]	陈家扬, 陈华, 张旭 . 基于 NWP 和深度学习神经网络短期风功率预测. 现代电子技术, 2020,43(8):63-67 (Chen Jiayang, Chen Hua, Zhang Xu. Short-term wind power forecasting based on NWP and deep learning neural network. Modern Electronics Technique. 2020,43(8):63-67 (in Chinese))
[20]	李江, 冯存前, 王义哲等. 基于深度卷积神经网络的弹道目标微动分类. 空军工程大学学报(自然科学版), 2019,20(4):97-102 (Li Jiang, Feng Cunqian, Wang Yizhe, et al. Micro-motion classification of ballistic targets based on deep convolutional neural network. Journal of Airforce Engineering University (Natural Science Edition). 2019,20(4):97-102 (in Chinese))
[21]	Wang C, Ma L, Li R. et al. Exploring trajectory prediction through machine learning methods. IEEE Access, 2019,7:101441-101452
[22]	Wang T, Leiter K, Plechac P. et al. Accelerated scale bridging with sparsely approximated Gaussian learning. Journal of Computational Physics, 2019,1:1-21
[23]	Wang Y, Cheung SW, Chung ET. et al. Deep multiscale model learning. Journal of Computational Physics, 2020,406:1-17
[24]	Yang X, Barajas-Solano D, Tartakovsky G. et al. Physics-informed cokriging: a gaussian-process-regression-based multifidelity method for data-model convergence. Journal of Computational Physics, 2019,395:410-431
[25]	Meng X, Karniadakis G. A composite neural network that learns from multi-fidelity data: Application to function approximation and inverse PDE problems. Journal of Computational Physics, 2020,401:1-15
[26]	周兵, 崔桂香, 陈乃祥. 基于尺度相似假设的大涡模拟动力方法. 清华大学学报: 自然科学版, 2006,46(8):1438-1441 (Zhou Bing, Cui Guixiang, Chen Naixiang. Dynamic procedure based on the scale-similarity hypotheses for large-eddy simulations. Journal of Tsinghua University (Science and Technology). 2006,46(8):1438-1441 (in Chinese))
[27]	傅霆. 医学信号与图像的多尺度分析方法研究. [博士论文]. 成都:电子科技大学, 2003 (Fu Ting. Multi-scale analysis research on biomedical signal and image processing. [PhD Thesis]. Chengdu: University of Electronic Science and Technology of China, 2003 (in Chinese))
[28]	冯志鹏, 朱萍玉, 褚福磊. 基于自适应多尺度线性调频小波分解的水轮机非平稳振动信号分析. 中国电机工程学报, 2008,28(8):107-112 (Feng Zhipeng, Zhu Pengyu, Chu Fulei. Time-frequency analysis of hydroturbine nonstationary vibration signal based on adaptive multi-scale chirplet decomposition. Proceedings of the CSEE. 2008,28(8):107-112 (in Chinese))
[29]	郭文璐, 李泓辰, 王静竹等. 单空泡与自由液面相互作用规律研究进展. 力学学报, 2019,51(6):1682-1698 (Guo Wenlu, Li Hongchen, Wang Jingzhu, et al. Reaserch progress on interaction between a single cavitation and free surface. Chinese Journal of Theoretical and Applied Mechanics. 2019,51(6):1682-1698 (in Chinese))
[30]	Alehossein H, Qin Z. Numerical analysis of Rayleigh-Plesset equation for cavitating water jets. International Journal for Numerical Methods in Engineering, 2007,72(7):780-807
[31]	Huang B, Qiu SC, Li XB. et al. A review of transient flow structure and unsteady mechanism of cavitating flow. Journal of Hydrodynamics, 2019,31(3):429-444
[32]	Xu H, Liu B, Shu L, et al. Double embeddings and {CNN}-based sequence labeling for aspect extraction//Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics, Melbourne, Australia, 2018, 592-598
[33]	叶舒然, 张珍, 宋旭东等. 自动编码器在流场降阶中的应用. 空气动力学学报, 2019,37(3):498-504 (Ye Shuran, Zhang Zhen, Song Xudong, et al. Applications of autoencoder in reduced-order modeling of flow field. Acta Aerodynamica Sinica. 2019,37(3):498-504 (in Chinese))