船舶尾流感应磁场信号较为微弱，目标回波信号与背景噪声的区分度低，导致回波信号检测的相对容限值较大。为此，提出船舶尾流感应磁场小目标微弱回波信号检测方法。分析尾流散射回波特性，结合尾流气泡分布尺寸获取船舶尾流整体散射回波信号。将整体散射回波信号输入长短期记忆循环神经网络中，检测微弱回波信号。测试结果显示，随着尾流传播距离的逐渐增加，磁场感应强度逐渐减弱；回波信号幅值低于0.4 V时，能够完成微弱回波信号检测；相对容限值均在0.15以下，满足回波信号检测需求。
The ship's wake induced magnetic field signal is weak, and the distinction between the target echo signal and the background noise is low, resulting in the relative tolerance limit of the echo signal detection is large. Therefore, a method for detecting weak echo signal of small target induced by ship wake magnetic field is proposed. The characteristics of wake scattering echo are analyzed, and the overall scatter echo signal of ship wake is obtained by combining with the distribution size of wake bubble. The whole scattered echo signal is input into the long short-term memory recurrent neural network to detect the weak echo signal. The test results show that the magnetic field induction intensity decreases with the increase of wake propagation distance. When the echo signal amplitude is lower than 0.4 V, the weak echo signal detection can be completed. The relative tolerance limits are all below 0.15, meeting the demand of echo signal detection.
2025,47(12): 106-110 收稿日期：2025-2-27
DOI：10.3404/j.issn.1672-7649.2025.12.019
分类号：TJ630
作者简介：刘旭东(1984-),男,硕士,讲师,研究方向为水下目标探测技术
参考文献：
[1] 宗思光, 张鑫, 曹静, 等. 舰船尾流激光探测跟踪方法与试验[J]. 红外与激光工程, 2023, 52(3): 205-216.
[2] 李想, 贾伟, 梁宁. 四桨船舶尾流场特性数值预报研究[J]. 船舶力学, 2024, 28(8): 1187-1199.
[3] 冯伦宇, 张志君, 李跃忠, 等. 基于静态峰值分布的超声波回波信号检测方法研究[J]. 中国测试, 2023, 49(1): 43-49.
[4] 苏映新. 基于优化匹配追踪稀疏分解的微弱超声回波提取[J]. 声学技术, 2023, 42(5): 616-620.
[5] 宋大雷, 孙兆阳, 孙康康, 等. 基于改进WOA-MP算法的多普勒海流计回波信号频率估计研究[J]. 中国测试, 2024, 50(8): 1–10.
[6] 许彦伟, 薛勐, 谷浩翔, 等. 耦合浅海声信道目标回波建模与检测性能分析[J]. 声学学报, 2024, 49(4): 719-730.
XU Y W, XUE M, GU H X, et al. Modeling and detection performance analysis of target echoes coupled with shallow water acoustic channels[J]. Acta Acustica, 2024, 49(4): 719-730.
[7] 宗思光, 段子科, 张鑫, 等. 复杂混合尾流气泡场的激光探测特性研究[J]. 半导体光电, 2023, 44(5): 741–746+781.
ZONG S G, DUAN Z K, ZHANG X, et al. Study on laser detection characteristics of complex mixed wake bubble field[J]. Semiconductor Optoelectronics, 2023, 49(1): 43–49.
[8] 徐东泽, 肖琴. 基于小波与注意力机制的金融序列BiLSTM预测[J]. 计算机仿真, 2024, 41(5): 527-531.
XU D Z, XIAO Q. BILSTM prediction of financial time series based on wavelet and attention mechanism[J]. Computer Simulation, 2024, 41(5): 527-531.
