潜艇隐蔽性是衡量其作战效能的重要指标，尤其在水声对抗条件下，准确预测其在海洋环境中的隐蔽概率对战术规避与路径规划具有关键意义。传统方法主要依赖声呐方程及声传播模型对信号余量进行计算，但此类方法对环境参数变化敏感，建模复杂且计算耗时，难以满足快速评估的需求。为此，本文提出一种融合物理建模与数据驱动的潜艇隐蔽性预测方法，首先以海洋环境参数为输入，通过 Bellhop模型获取传播损失，并结合主动声呐方程计算信号余量，随后构建多层感知器神经网络模型，以环境参数为输入特征、信号余量为预测目标，实现对信号余量的快速拟合。最终将预测结果代入隐蔽概率模型，完成对潜艇在多变环境下隐蔽性的评估。仿真表明，该方法不仅在预测精度上与传统模型保持一致，还提升了计算效率，验证了其在复杂水声环境中的可行性与应用价值。
The concealment of submarines is a crucial indicator for evaluating their combat effectiveness, especially under conditions of underwater acoustic countermeasures. Accurately predicting the probability of concealment in marine environments is of vital significance for tactical evasion and path planning. Traditional methods mainly rely on sonar equations and acoustic propagation models to calculate the signal margin, but these methods are sensitive to changes in environmental parameters, have complex modeling, and are computationally time-consuming, making it difficult to meet the requirements of rapid assessment. Therefore, this paper proposes a submarine concealment prediction method that integrates physical modeling and data-driven approaches. Firstly, the Bellhop model is used to obtain the propagation loss with marine environmental parameters as input, and the signal margin is calculated in combination with the active sonar equation. Then, a multi-layer perceptron neural network model is constructed, with environmental parameters as input features and the signal margin as the prediction target, to achieve rapid fitting of the signal margin. Finally, the predicted signal margin is incorporated into a concealment probability model to assess submarine stealth performance across varying environmental conditions. The simulation results show that this method not only maintains the prediction accuracy of traditional models but also improves computational efficiency, verifying its feasibility and application value in complex underwater acoustic environments.
2026,48(5): 93-102 收稿日期：2025-7-2
DOI：10.3404/j.issn.1672-7649.2026.05.015
分类号：U676.1；TP3
基金项目：国家自然科学基金资助项目（72371137）
作者简介：黄磊(2000-),男,硕士研究生,研究方向为应急管理
参考文献：
[1] 韩小溪, 刘洋. 美国核潜艇装备技术未来发展研究[J]. 舰船科学技术, 2020, 42(19): 179-181
HAN X X, LIU Y. Research on the future development of U. S. nuclear submarine equipment technology[J]. Ship Science and Technology, 2020, 42(19): 179-181
[2] 仝博, 李永清, 张振海, 等. 复合材料多层圆柱壳振动和声辐射问题研究进展[J]. 材料导报, 2023, 37(2): 256-263
TONG B, LI Y Q, ZHANG Z H, et al. Research advances on the vibration and sound radiation of composite multilayer cylindrical shells[J]. Materials Reports, 2023, 37(2): 256-263
[3] 胡潇, 向阳, 王文博, 等. 双壳体潜艇流水孔对流激噪声的影响分析[J]. 舰船科学技术, 2025, 47(1): 124-132
HU X, XIANG Y, WANG W B, et al. Analysis of flow-induced noise influence by water holes in double-hull submarines[J]. Ship Science and Technology, 2025, 47(1): 124-132
[4] LIU J W, YANG H B, ZHAO H G, et al. A lightweight waterborne acoustic meta-absorber with low characteristic impedance rods[J]. International Journal of Mechanical Sciences, 2023, 255: 108469
[5] 过武宏, 范培勤, 刘敬一, 等. 黄海夏季温跃层适应性观测对声传播的影响[J/OL]. 应用声学, 1–9.
GUO W H, FAN P Q, LIU J Y, et al. Influence of adaptive observation information of the summer thermocline on sound propagation in the Yellow Sea[J/OL]. Journal of Applied Acoustics, 1–9.
[6] 秦锋, 赵志允, 于振涛. 非均匀温盐场潜艇声隐蔽效能仿真研究[J]. 青岛大学学报(工程技术版), 2020, 35(4): 62-67
QIN F, ZHAO Z Y, YU Z T. Simulation of submarine's acoustic concealment effectiveness in heterogeneous temperature salt field[J]. Journal of Qingdao University (E& T), 2020, 35(4): 62-67
[7] GUO Z L, HONG M, SHI J, et al. Risk assessment of unmanned underwater platforms in the South China Sea marine environment based on two-dimensional density-weighted operator and cloud barycentric Bayesian combination[J]. Ocean Engineering, 2023, 283: 115184.
[8] 汪晶晗, 陈欢, 金宇琦, 等. 一种兼容海洋环境的改进Transformer声呐探测效能快速预报模型[J]. 声学技术, 2025, 44(2): 164-170
WANG J H, CHEN H, JIN Y Q, et al. An improved Transformer model for rapid prediction of sonar detection performance compatible with marine environment[J]. Technical Acoustics, 2025, 44(2): 164-170
[9] NIU H Q, REEVES E, GERSTOFT P. Source localization in an ocean waveguide using supervised machine learning[J]. The Journal of the Acoustical Society of America, 2017, 142(3): 1176-1188
[10] 王凯, 许昭霞, 李岳阳, 等. 美国海军海洋测量船发展及使用研究[J]. 舰船科学技术, 2020, 42(19): 185-189
WANG K, XU Z X, LI Y Y, et al. Research on the development and use of the U. S. navy oceanographic research ships[J]. Ship Science and Technology, 2020, 42(19): 185-189
[11] 何心怡, 蔡志明, 林建域, 等. 主动声呐探测距离预报仿真研究(英文)[J]. 系统仿真学报, 2003(9): 1304-1306+1310
HE X Y, CAI Z M, LIN J Y, et al. The simulation research on forecasting the detection range of active sonar[J]. Journal of System Simulation., 2003(9): 1304-1306+1310
[12] URICK R J. Principles of underwater sound /-3rd ed[M]. McGraw-Hill Book Company, 1983.
[13] WILSON W D. Equation for the speed of sound in sea water[J]. The Journal of the Acoustical Society of America, 1960, 32(10): 1357-1357
[14] DEL GROSSO V A. New equation for the speed of sound in natural waters (with comparisons to other equations)[J]. The Journal of the Acoustical Society of America, 1974, 56(4): 1084-1091
[15] MACKENZIE K V. Nine‐term equation for sound speed in the oceans[J]. The Journal of the Acoustical Society of America, 1981, 70(3): 807-812
[16] CHEN C T, MILLERO F J. Speed of sound in seawater at high pressures[J]. The Journal of the Acoustical Society of America, 1977, 62(5): 1129-1135
[17] ETTER P C. Underwater acoustic modeling and simulation[M]. CRC Press, 2018.
[18] THORP W H. Analytic description of the low-frequency attenuation coefficient[J]. The Journal of the Acoustical Society of America, 1967, 42(1): 270-270
[19] 刘伯胜, 雷家煜. 水声学原理[M]. 哈尔滨: 哈尔滨工程大学出版社, 2010.
[20] AINSLIE M A. Principles of sonar performance modelling[M]. Berlin: Springer, 2010.
[21] 冯志奇, 司光亚. 基于概率计算的航母编队探潜能力评估[J]. 指挥控制与仿真, 2022, 44(2): 63-70
FENG Z Q, SI G Y. Evaluation of submarine detection capability of aircraft carrier formation based on probability calculation[J]. Command Control & Simulation, 2022, 44(2): 63-70
[22] HASTIE T, TIBSHIRANI R, FRIEDMAN J. The elements of statistical learning: data mining, inference, and prediction [M]. 2nd ed. New York: Springer, 2009.
[23] 屈可扬, 程静, 甘云华, 等. 基于时序数据集划分和时序交叉验证优化燃煤锅炉NOx建模[J]. 中南大学学报(自然科学版), 2024, 55(12): 4665-4674
QU K Y, CHENG J, GAN Y H, et al. Optimization of NOx modeling for coal-fired boilers based on time series dataset division and time series cross-validation[J]. Journal of Central South University (Science and Technology), 2024, 55(12): 4665-4674
