基于RLPSO算法的卷积神经网络地震P波识别方法研究*

doi:10.11978/2025041

热带海洋学报 ›› 2025, Vol. 44 ›› Issue (6): 31-38.doi: 10.11978/2025041CSTR: 32234.14.2025041

基于RLPSO算法的卷积神经网络地震P波识别方法研究^*

王罗文¹^,²^,³(), 杨泽荣⁴, 文永鹏⁴, 朱心科²^,³, 颜波¹^,²^,³, 秦华伟¹()

¹ 杭州电子科技大学机械工程学院, 浙江杭州 310018
² 海底科学与划界全国重点实验室, 浙江杭州 310012
³ 自然资源部第二海洋研究所, 浙江杭州 310012
⁴ 国电电力浙江舟山海上风电开发有限公司, 浙江舟山 316000

收稿日期:2025-03-11 修回日期:2025-04-11 出版日期:2025-11-10 发布日期:2025-12-03
通讯作者: 秦华伟。email: qinhw@hdu.edu.cn
作者简介:
王罗文(2000—), 男, 黑龙江省黑河市人, 硕士研究生, 主要研究方向为海洋机电装备技术。email: t2777269216@163.com
基金资助:
国家重点研发计划(2022YFC3003802); 自然资源部海洋观测技术重点实验室研究基金(2024klootA11); 自然资源部海底科学重点实验室开放基金(KLSG2409)

Research on seismic P-wave identification method based on convolutional neural network optimized by RLPSO algorithm^*

WANG Luowen¹^,²^,³(), YANG Zerong⁴, WEN Yongpeng⁴, ZHU Xinke²^,³, YAN Bo¹^,²^,³, QIN Huawei¹()

¹ School of Mechanical Engineering, Hangzhou Dianzi University, Hangzhou 310018, China
² State Key Laboratory of Submarine Geoscience, Hangzhou 310012, China
³ Second Institute of Oceanography, MNR, Hangzhou 310012, China
⁴ Guodian Electric Power Zhejiang Zhoushan Offshore WindPower Development Co., Ltd., Zhoushan 316000, China

Received:2025-03-11 Revised:2025-04-11 Online:2025-11-10 Published:2025-12-03
Contact: QIN Huawei. email: qinhw@hdu.edu.cn
Supported by:
National Key R&D Program of China(2022YFC3003802); Research Fund of the Key Laboratory of Marine Observation Technology, Ministry of Natural Resources(2024klootA11); Open Fund of the Key Laboratory of Submarine Geoscience, Ministry of Natural Resources(KLSG2409)

摘要/Abstract

摘要：

针对海洋地震P波实时监测系统的地震P波识别需求, 设计了一种基于深度学习的地震P波识别算法。该算法通过构建卷积神经网络和强化学习粒子群优化算法的混合地震P波识别模型, 利用先验地震数据对模型进行训练, 并采用RLPSO (reinforcement learning particle swarm optimization)算法对卷积神经网络(convolutional neural network, CNN) 模型中的超参数进行迭代优化。将优化后的超参数代入CNN模型后, 实现了地震P波的识别。为了验证算法的有效性, 采用测试数据集对不同算法进行了仿真测试。实验结果表明, 该算法在训练过程中表现出更低的训练损失和更高的识别准确率。此外, 该算法在低信噪比条件下仍能保持较高的识别精度, 展现了较强的鲁棒性, 满足了海洋地震P波实时监测系统的地震P波识别需求。

关键词: 地震P波识别, 卷积神经网络, 强化学习, 粒子群优化算法

Abstract:

In response to the demand for real-time monitoring of seismic P-waves in marine environments, this study designs a deep learning-based algorithm for seismic P-wave identification. The algorithm constructs a hybrid model by integrating a convolutional neural network (CNN) with a reinforcement learning particle swarm optimization (RLPSO) algorithm. The model is trained using prior seismic data as prior information, with the RLPSO algorithm iteratively optimizing the hyperparameters of the CNN. The optimized hyperparameters are then incorporated into the CNN model to predict the arrival time of seismic P-waves. To validate the effectiveness of the algorithm, simulation tests were conducted on a test dataset using different algorithms. The experimental results demonstrate that the proposed algorithm exhibits lower training loss and higher identification accuracy during the training. Furthermore, the algorithm maintains high identification precision under low signal-to-noise ratio conditions, showcasing strong robustness. The algorithm meets the requirements for real-time seismic P-wave identification in marine seismic monitoring systems.

Key words: seismic P-wave identification, convolutional neural network, reinforcement learning, particle swarm optimization algorithm

中图分类号:

P738.4

王罗文, 杨泽荣, 文永鹏, 朱心科, 颜波, 秦华伟. 基于RLPSO算法的卷积神经网络地震P波识别方法研究^*[J]. 热带海洋学报, 2025, 44(6): 31-38.

WANG Luowen, YANG Zerong, WEN Yongpeng, ZHU Xinke, YAN Bo, QIN Huawei. Research on seismic P-wave identification method based on convolutional neural network optimized by RLPSO algorithm^*[J]. Journal of Tropical Oceanography, 2025, 44(6): 31-38.

图/表 8

图1

表1

图2

图3

表2

图4

表3

图5

参考文献 11

[1]	丁巍伟, 黄豪彩, 朱心科, 等, 2019. 一种新型潜标式海洋地震仪及在海洋地震探测中的应用[J]. 地球物理学进展, 34(1): 292-296.
	DING WEIWEI, HUANG HAOCAI, ZHU XINKE, et al, 2019. New mobile oceanic seismic recording system and its application in marine seismic exploration[J]. Progress in Geophysics, 34(1): 292-296 (in Chinese with English abstract).
[2]	张军华, 赵勇, 赵爱国, 等, 2002. 用小波变换与能量比方法联合拾取初至波[J]. 物探化探计算技术, 24(4): 309-312+336.
	ZHANG JUNHUA, ZHAO YONG, ZHAO AIGUO, et al, 2002. Seismic first break pickup using wavelet transform and power ratio method[J]. Computing Techniques for Geophysical and Geochemical Exploration, 24(4): 309-312+336 (in Chinese with English abstract).
[3]	ALLEN R V, 1978. Automatic earthquake recognition and timing from single traces[J]. Bulletin of the Seismological Society of America, 68(5): 1521-1532. doi: 10.1785/BSSA0680051521
[4]	KENNEDY J, EBERHART R, 1995. Particle swarm optimization[C]// Proceedings of ICNN'95 - International Conference on Neural Networks. Perth, WA, Australia: IEEE: 1942-1948.
[5]	KHALIFA M H, AMMAR M, OUARDA W, et al, 2017. Particle swarm optimization for deep learning of convolution neural network[C]//2017 Sudan Conference on Computer Science and Information Technology (SCCSIT). Elnihood, Sudan: IEEE: 1-5.
[6]	MAEDA N, 1985. A method for reading and checking phase time in auto-processing system of seismic wave data[J]. Zisin (Journal of the Seismological Society of Japan 2nd Ser), 38(3): 365-379.
[7]	MOUSAVI S M, SHENG YIXIAO, ZHU WEIQIANG, et al, 2019. STanford EArthquake Dataset (STEAD): a global data set of seismic signals for AI[J]. IEEE Access, 7: 179464-179476. doi: 10.1109/Access.6287639
[8]	RAMOS N, LI HAI, 2024. CNN implementation of Bayesian plasticity for robust learning[C]//2024 IEEE 6th International Conference on AI Circuits and Systems (AICAS). Abu Dhabi, United Arab Emirates: IEEE: 432-436.
[9]	ROBERTS R G, CHRISTOFFERSSON A, CASSIDY F, 1989. Real-time event detection, phase identification and source location estimation using single station three-component seismic data[J]. Geophysical Journal International, 97(3): 471-480. doi: 10.1111/gji.1989.97.issue-3
[10]	WANG QIANG, ZHAN ZHONGLI, 2011. Reinforcement learning model, algorithms and its application[C]// 2011 International Conference on Mechatronic Science, Electric Engineering and Computer (MEC). Jilin, China: IEEE: 1143-1146.
[11]	WOOLLAM J, RIETBROCK A, BUENO A, et al, 2019. Convolutional neural network for seismic phase classification, performance demonstration over a local seismic network[J]. Seismological Research Letters, 90(2A): 491-502. doi: 10.1785/0220180312

方法	准确率	均方根误差/s	漏检率
CNN	83.88%	0.3797	10.74%
PSO-CNN	91.36%	0.2974	7.06%
RLPSO-CNN	95.90%	0.1660	3.26%

方法	准确率	均方根误差/s	漏检率	训练时长/h
RLPSO-CNN	95.90%	0.1660	3.26%	20.1
Bayesian-CNN	82.21%	0.4739	12.10%	22.5

基于RLPSO算法的卷积神经网络地震P波识别方法研究^*

Research on seismic P-wave identification method based on convolutional neural network optimized by RLPSO algorithm^*

RichHTML

PDF (PC)

赞

可视化

摘要/Abstract

引用本文

使用本文

图/表 8

参考文献 11

相关文章 4

Metrics

本文评价

推荐阅读 0

[1]	许毅, 韦骏, 魏春雷, 杨凡. 基于CNN-LSTM的高频地波雷达海流方向校正方法[J]. 热带海洋学报, 2025, 44(3): 24-35.
[2]	王傲宇, 屈科, 王旭, 高榕泽, 门佳. 基于非静压模型数值模拟与卷积神经网络的滑坡涌浪水动力特性预测[J]. 热带海洋学报, 2025, 44(2): 187-195.
[3]	高榕泽, 屈科, 任兴月, 王旭. 卷积神经网络方法在岛礁类海啸波水动力特性演变的应用[J]. 热带海洋学报, 2024, 43(4): 68-75.
[4]	翁少佳, 蔡锦海, 庞运禧, 罗荣真. 卷积神经网络在近岸表层海温预报中的应用[J]. 热带海洋学报, 2024, 43(1): 40-47.

输入: 地震P波验证集$D=\left\{ \left( {{X}_{k}},{{Y}_{k}} \right)\|k=1,2,...,m \right\}$、地震P波训练集 ${{D}_{\text{train}}}=\left\{ \left( X_{v}^{T},Y_{v}^{T} \right)\|v=1,2,...,w \right\}$、地震P波测试集 ${{D}_{\text{test}}}=\left\{ \left( X_{u}^{},Y_{u}^{} \right)\|u=1,2,...,n \right\}$、最大迭代次数T、粒子群数目N、初始参数w(t) 输出: ${{P}_{\text{out}}}=$ (全局最优参数组合$gbest$, 最小验证集损失$gbest\_score$, 正确率$accuracy$)
1. 在边界范围内初始化粒子群, 随机初始化粒子位置和速度
2. for $t=0,...,T$
3. {
4. for $i=0,...,N$
5. {
6. $model=EnhancedPWaveCNN\left( HyperParameter \right)$ //解析超参数构建CNN模型
7. $val\_loss=\frac{1}{m}\sum\limits_{j=1}^{m}{SmoothL1Loss\left( model\left( {{X}_{j}} \right),{{Y}_{j}} \right)}$ //计算验证损失
8. if $val\_loss<gbest\_score$
9. $gbest=particles[i],gbest\_score=val\_loss$
10. else $particles[i]=particles[i-1]$/更新全局最优解和最优损失
11. $velocities[i]=w\cdot velocities[i]+{{c}_{1}}\cdot rand()\cdot (pbest[i]-particles[i])+{{c}_{2}}\cdot rand()\cdot (gbest-particles[i])$
12. $particles[i]+=velocities[i]$//更新粒子位置和速度
13. $w\left( t+1 \right)=w\left( t \right)+\Delta w\left( t \right)$ //调整惯性权重和加速常数, 计算奖励
14. }
15. }
16. $train\_model\left( model,gbest,{{D}_{\text{train}}} \right)$
17. ${{P}_{out}}=test\_model(model,{{D}_{\text{test}}})$ //以最优超参数训练和测试输出结果

基于RLPSO算法的卷积神经网络地震P波识别方法研究*

Research on seismic P-wave identification method based on convolutional neural network optimized by RLPSO algorithm*

RichHTML

PDF (PC)

赞

可视化

摘要/Abstract

引用本文

使用本文

图/表 8

参考文献 11

相关文章 4

Metrics

本文评价

推荐阅读 0

基于RLPSO算法的卷积神经网络地震P波识别方法研究^*

Research on seismic P-wave identification method based on convolutional neural network optimized by RLPSO algorithm^*