[1]周明胡,赵英亮,韩星程,等.基于 Bi-LSTM-MLP 的水下声速预测[J].机械与电子,2025,(11):8-14.
　ZHOU Minghu,ZHAO Yingliang,HAN Xingcheng,et al.Underwater Sound Speed Prediction Based on Bi-LSTM-MLP[J].Machinery & Electronics,2025,(11):8-14.
点击复制

基于 Bi-LSTM-MLP 的水下声速预测()
分享到：

《机械与电子》[ISSN:1001-2257/CN:52-1052/TH]

卷:
期数:
2025年11期
页码:
8-14
栏目:
研究与设计
出版日期:
2025-11-24

文章信息/Info

Title:
Underwater Sound Speed Prediction Based on Bi-LSTM-MLP
文章编号:
1001-2257 ( 2025 ) 11-0008-07
作者:
周明胡; 赵英亮; 韩星程; 黄嘉瑒; 王国祥
中北大学信息与通信工程学院,山西 太原 030051
Author(s):
ZHOU Minghu ;  ZHAO Yingliang ;  HAN Xingcheng ;  HUANG Jiayang ;  WANG Guoxiang
( School of Information and Communication Engineering , North University of China , Taiyuan 030051 , China )
关键词:
声速预测; 深度学习; Bi-LSTM-MLP ;  SVP
Keywords:
sound speed prediction ;  deep learning ;  Bi-LSTM-MLP ;  SVP
分类号:
TP18 ;TB56
文献标志码:
A
摘要:
针对直接测量获取声速剖面( SVP )数据设备成本高、频繁观测效率低下的问题,提出一种基于双向长短期记忆网络与多层感知机融合模型( Bi-LSTM-MLP )的 SVP 预测方法。该模型融合了双向长短期记忆网络( Bi-LSTM )对时序前后依赖关系的建模能力与多层感知机( MLP )在非线性特征映射和空间整合上的优势,有效提升了 SVP 预测的精度与泛化性能。所提方法在相同条件下相较于 Conv-LSTM 和Bi-LSTM 模型,在 MSE 指标上分别提升了约 43% 和 74% 。在 6 个不同地理位置下的预测结果中,模型的准确率指标则基本稳定在 99.91% 以上;而在固定位置的月度连续预测中, MSE 最大值控制在 0.50 m / s ,验证了模型在时空维度上的准确率与鲁棒性。
Abstract:
To address the issues of high equipment cost for direct measurement of sound velocity profile ( SVP ) data and low efficiency of frequent observations , a sound velocity profile ( SVP ) prediction method based on a fusion model of bidirectional long short-term memory network and multi-layer perceptron ( Bi-LSTM-MLP ) is proposed.This model combines the bidirectional long short-term memory network ( Bi LSTM )’ s ability to model the dependencies in both directions of time series with the multi layer perceptron ( MLP )’ s advantages in nonlinear feature mapping and spatial integration , effectively enhancing the accuracy and generalization performance of sound velocity profiles prediction.Compared with the Conv-LSTM and Bi-LSTM models under the same conditions , the proposed method improves the mean squared error ( MSE ) by approximately 43% and 74% respectively.In the prediction results from six different geographical locations , the model ’ s accuracy metric remained stable above 99.91% ; while in the monthly continuous prediction at a fixed location , the maximum MSE is controlled at around 0.50 m / s , verifying the model ’ s accuracy and robustness in both temporal and spatial dimensions.

参考文献/References:

[ 1 ] 冯业,杜民 . 基于无人机平台的水下通信中继技术研究[ J ] . 电脑与信息技术, 2025 , 33 ( 2 ): 61-63 , 90.

[ 2 ] LU J J , ZHANG H , LI S J , et al.Enhancing few-shot prediction of ocean sound velocity profiles through hierarchical long short-term memory transfer learning[ J ] .Journal of marine science and engineering , 2024 , 12( 7 ): 1041.
[ 3 ] 黄威,高凡,王君婷,等 . 水下声速场构建方法综述[ J ] .哈尔滨工程大学学报,2023 , 44 ( 11 ): 2005-2017.
[ 4 ] HUANG W , LI D S , JIANG P.Underwater sound speed inversion by joint artificial neural network and ray theory [ C ] ∥13th International Conference on Underwater Networks and Systems , 2018 : 1-8.
[ 5 ] LU J J , HUANG W , ZHANG H.Dynamic prediction of full-ocean depth SVP by a hierarchical LSTM : an experimental result [ J ] .IEEE Geoscience and remote sensing letters , 2024 , 21 : 1501105.
[ 6 ] GANGOPADHYAY T , TAN S Y , JIANG Z H , et al. Spatiotemporal attention for multivariate time series prediction and interpretation [ C ] ∥ICASVP 2021 2021 IEEE International Conference on Acoustics , Speech and Signal Processing ( ICASVP ) .New York : IEEE , 2021 : 3560-3564.
[ 7 ] LI B Y , ZHAI J S.A novel sound velocity profile prediction method based on the convolutional long-short term memory network [ J ] .Journal of marine science and engineering , 2022 , 10 ( 5 ): 572.
[ 8 ] YU Y , SI X S , HU C H , et al.A review of recurrent neural networks : LSTM cells and network architectures [ J ] . Neural computation , 2019 , 31 ( 7 ): 1235-1270.
[ 9 ] SHIRI M F , PERUMAL T , MUSTAPHA N , et al.A comprehensive overview and comparative analysis on deep learning models [ J ] .Journal on artificial intelligence , 2024 , 6 ( 1 ): 301-360.
[ 10 ] LI Z L , YU H Y , XU J X , et al.Stock market analysis and prediction using LSTM : a case study on technology stocks [ J ] .Innovations in applied engineering and technology , 2023 , 2 ( 1 ): 1-6.
[ 11 ] SIAMI-NAMINI S , TAVAKOLI N , NAMIN A S. The performance of LSTM and BiLSTM in forecasting time series [ C ] ∥2019 IEEE International Conference on Big Data ( Big Data ) .New York : IEEE , 2019 : 3285-3292.
[ 12 ] WO?NIAK M , WIECZOREK M , SIKA J.BiLSTM deep neural network model for imbalanced medical data of IoT systems [ J ] .Future generation computer systems , 2023 , 141 : 489-499.
[ 13 ] HAMEED Z , GARCIA-ZAPIRAIN B.Sentiment classification using a single-layered BiLSTM model [ J ] . IEEE Access , 2020 , 8 : 73992-74001.
[ 14 ] 郑宇鸿,徐佳毅,文艺成,等 . 基于多层感知机的远距离水下 声 源 定 位 技 术 研 究 [ J ] . 导 航 定 位 与 授 时,2022 , 9 ( 3 ): 40-48.
[ 15 ] MEINEN C S , WATTS D R.Further evidence that the sound-speed algorithm of Del Grosso is more accurate than that of Chen and Millero [ J ] .The journal of the acoustical society of America , 1997 , 102 ( 4 ):2058-2062.

相似文献/References:

[1]王骁贤,张保华,陆思良.基于连续小波变换和卷积神经网络的无刷直流电机故障诊断[J].机械与电子,2018,(06):29.
　WANG Xiaoxian,ZHANG Baohua,LU Siliang.Fault Diagnosis of Brushless Direct Current Motor Based on Continuous Wavelet Transform and Convolutional Neural Network[J].Machinery & Electronics,2018,(11):29.
[2]刘志宇,黄亦翔.基于深度学习和迁移学习的液压泵健康评估方法[J].机械与电子,2018,(09):67.
　LIU Zhiyu,HUANG Yixiang.Health Assessment for Hydraulic Pump Based on Deep Learning and Transfer Learning[J].Machinery & Electronics,2018,(11):67.
[3]肖倩宏,康 鹏,杜 江,等.深度学习在电网智能调控系统中应用研究[J].机械与电子,2021,(01):38.
　XIAO Qianhong,KANG Peng,DU Jiang,et al.Research on the Application of Deep Learning Theory in Power Grid Intelligent Dispatching[J].Machinery & Electronics,2021,(11):38.
[4]许 哲,张少帅,郭 璐,等.无人机深度学习去雾算法[J].机械与电子,2021,(04):13.
　XU Zhe,ZHANG Shaoshuai,GUO Lu,et al. Deep Learning Defogging Algorithm for UAV[J].Machinery & Electronics,2021,(11):13.
[5]齐爱玲１,李 琳１,朱亦轩２,等.基于融合特征的双通道CNN滚动轴承故障识别[J].机械与电子,2021,(05):15.
　QI Ailing,LI Lin,ZHU Yixuan,et al.Dual Channel CNN Bearing Fault Identification Based on Fusion Feature[J].Machinery & Electronics,2021,(11):15.
[6]徐先峰,郑少杰,赵 依,等.基于数据分解与重构的光伏发电功率超短期预测[J].机械与电子,2022,(04):20.
　XU Xianfeng,ZHENG Shaojie,ZHAO Yi,et al.Ultra-short-term Prediction of Photovoltaic Power Generation Based on Data Decomposition and Deconstruction[J].Machinery & Electronics,2022,(11):20.
[7]江 励,熊达明,汤健华,等.自然光线环境中的空间物体快速识别和定位算法研究[J].机械与电子,2022,(06):8.
　JIANG Li,XIONG Daming,TANG Jianhua,et al.Recognition and Positioning Algorithm of Space Objects in Natural Light Environment[J].Machinery & Electronics,2022,(11):8.
[8]王西志,管声启,张理博,等.基于视觉引导的工业棒材上料系统研究[J].机械与电子,2023,41(05):19.
　WANG Xizhi,GUAN Shengqi,ZHANG Libo,et al.Research on Industrial Bar Feeding System Based on Visual Guidance[J].Machinery & Electronics,2023,41(11):19.
[9]王 青,吕绪山,党 帅,等.基于深度学习的纱管识别方法研究[J].机械与电子,2023,41(12):20.
　WANG Qing,LYU Xushan,DANG Shuai,et al.Research on Yarn Bobbin Detection Method Based on Deep Learning[J].Machinery & Electronics,2023,41(11):20.
[10]姜越夫,王 青,吕绪山.改进 YOLOv5s 的纱管目标检测方法[J].机械与电子,2024,42(02):29.
　JIANG Yuefu,WANG Qing,LYU Xushan.Improved YOLOv5s Method for Yarn Tube Object Detection[J].Machinery & Electronics,2024,42(11):29.

备注/Memo

备注/Memo:
收稿日期: 2025-07-01
基金项目:国家自然科学青年基金资助项目( 62203405 );山西省应用基础研究计划项目( 202303021212206 );山西省重点研发计划项目( 202202110401015 )
作者简介:周明胡 ( 2001- ),男,安徽蚌埠人,硕士研究生,研究方向为水下通信与定位;赵英亮 ( 1975- ),女,山西运城人,博士,副教授,硕士研究生导师,研究方向为信号处理,通信作者, E-mail : zbhanxc@nuc.edu.cn
更新日期/Last Update: 2025-12-10