针对含内部结构的水下薄壳振动声辐射问题提出一种快速预报方法。该方法采用贝叶斯超参数优化方法构建神经网络模型，建立薄壳结构表面振动响应与声压的映射关系。通过与理论模拟和实验结果对比验证了该方法的可靠性。与实验结果对比表明，该方法在1～100 Hz范围内，峰值处的平均绝对误差为6.31 dB；与有限元法计算结果相比，总声压级的拟合优度为97.79%，平均绝对误差为0.85 dB。该方法具有较高的计算效率，可为水下结构振动声辐射快速预报提供新思路。
A fast prediction method for acoustic radiation of underwater thin shell with internal structure is proposed. The neural network model is constructed with Bayesian hyperparameter optimization method, and the mapping relationship between surface vibration response and sound pressure is established. The reliability of the proposed method is verified by comparison with theoretical simulation and experimental results. The comparison with experimental results shows that the mean absolute error at the peak value is 6.31 dB in the range of 1～100 Hz. Compared with the finite element method, the goodness of fit of the total sound pressure level is 97.79% and the average absolute error is 0.85 dB. This method has high computational efficiency and provides a new idea for rapid prediction of acoustic radiation of underwater structure vibration.
2024,46(24): 35-39 收稿日期：2024-4-18
DOI：10.3404/j.issn.1672-7649.2024.24.006
分类号：U663.1
基金项目：国家自然科学基金资助项目（12072298）
作者简介：孙滨(1999-),男,硕士研究生,研究方向为水下结构振动声辐射预报
参考文献：
[1] 章林柯, 崔立林. 潜艇机械噪声源分类识别的小样本研究思想及相关算法评述[J]. 船舶力学, 2011, 15(8): 940-947.
[2] 李天匀, 刘敬喜, 周健. 基于格林函数的无障板结构声辐射分析[J]. 华中科技大学学报(自然科学版), 2003, 31(4): 99-101.
[3] 张磊, 马逊军, 鲁民月, 等. 一种含内部结构的水下圆柱壳振动声辐射计算方法[J]. 舰船科学技术, 2021, 43(12): 122-127.
ZHANG L, MA X J, LU M Y, et al. A method for vibration and sound radiation analysis of an immersed cylindrical shell with internal structures[J]. Ship Science and Technology, 2021, 43(12): 122-127.
[4] YANG H, SEONG W. Acoustic radiation efficiency of a submerged periodic ring-stiffened cylindrical shell with finite vibration loading[J]. Applied Acoustics, 2020, 171: 107664.
[5] 张阳阳, 楼京俊, 朱石坚. 弹性板-圆柱壳水下声辐射特性及影响因素[J]. 舰船科学技术, 2016, 38(11): 44-47.
ZHANG Y Y, LOU J J, ZHU S J. Underwater acoustic radiation characteristics and influencing factors of elastic plate-cylindrical shell[J]. Ship Science and Technology, 2016, 38(11): 44-47.
[6] 何其伟, 李海峰, 俞翔. 单/双层水下结构振动声辐射研究[J]. 舰船科学技术, 2017, 39(1): 17-20.
HE Q W, LI H F, YU X. Research on the vibration and acoustic radiation of single and double layers underwater structure[J]. Ship Science and Technology, 2017, 39(1): 17-20.
[7] 王升贵, 胡桥, 陈迎亮, 等. 基于深度学习的水下目标识别方法研究[J]. 舰船科学技术, 2020, 42(12): 141-145.
WANG S G, HU Q, CHEN Y L, et al. Research on underwater target recognition method based on deep learning[J]. Ship Science and Technology, 2020, 42(12): 141-145.
[8] 张艳宁, 焦李成, 靳云姬, 等. 一种基于自适应高斯神经网络的船舶噪声分类方法[J]. 声学学报, 1998, 23(4): 349-356.
[9] 周金牛, 李亚安, 吴岳松. 水声信号预测的神经网络方法[J]. 声学技术, 2006, 25(3): 226-229.
[10] OH B K, GLISIC B, PARK H S. Neural network-based seismic response prediction model for building structures using artificial earthquakes[J]. Journal of Sound and Vibration, 2020, 468: 22-46.
[11] SHEN X, HAO Q W, XIONG G. Modelling and predictive investigation on the vibration response of a propeller shaft based on a convolutional neural network[J]. Mechanical Sciences, 2022, 13(1): 485-494.
[12] 祝青鑫, 王浩, 茅建校, 等. 基于ANFIS的环境激励下桥梁结构应变响应预测分析[J]. 中国公路学报, 2019, 32(11): 62-70+117.
[13] SNOEK J, LAROCHELLE H, ADAMS R P. Practical bayesian optimization of machine learning algorithms [J]. CoRR, 2012, 12(6): 29-44.
[14] YANG L, SHAMI A. On hyperparameter optimization of machine learning algorithms: Theory and practice[J]. Neurocomputing, 2020, 415: 295-316.
[15] CHEN L, LI P, CHEN H R. Vibro-acoustic analysis of an underwater cylindrical shell with internal structures and a comparison with experiment[J]. Journal of Modern Mechanical Engineering and Technology, 2023, 10(5): 161-176.
