基于长短期记忆神经网络的板裂纹损伤检测方法

doi:10.16183/j.cnki.jsjtu.2020.095

上海交通大学学报 ›› 2021, Vol. 55 ›› Issue (5): 527-535.doi: 10.16183/j.cnki.jsjtu.2020.095

所属专题：《上海交通大学学报》2021年12期专题汇总专辑；《上海交通大学学报》2021年“交通运输工程”专题

基于长短期记忆神经网络的板裂纹损伤检测方法

张松林, 马栋梁, 王德禹()

高新船舶与深海开发装备协同创新中心, 上海 200240

收稿日期:2020-04-01 出版日期:2021-05-28 发布日期:2021-06-01
通讯作者: 王德禹 E-mail:dywang@sjtu.edu.cn
作者简介:张松林(1995-),男,山东省济宁市人,硕士生,主要研究方向为船体结构损伤检测.
基金资助:
工信部高技术船舶科研项目([2016]548);教育部财政部重大专项(201335)

Method for Plate Crack Damage Detection Based on Long Short-Term Memory Neural Network

ZHANG Songlin, MA Dongliang, WANG Deyu()

Collaborative Innovation Center for Advanced Ship and Deep-Sea Exploration, Shanghai Jiao Tong University, Shanghai 200240, China

Received:2020-04-01 Online:2021-05-28 Published:2021-06-01
Contact: WANG Deyu E-mail:dywang@sjtu.edu.cn

摘要/Abstract

摘要：

针对板不同位置裂纹损伤的智能分类问题,提出了一种基于长短期记忆(LSTM)神经网络的板裂纹损伤检测方法.采用Abaqus二次开发建立板裂纹损伤模型,计算高斯白噪声激励下板的加速度响应,并通过数据扩充方法生成数据集,同时考虑了噪声对损伤检测的影响.建立基于LSTM的板裂纹智能检测模型,直接将板的加速度响应作为输入,不需要额外的损伤特征提取,并以最小预测误差为目标,选择模型的超参数,优化模型配置.与多层感知机模型和基于小波包变换的多层感知机模型进行对比表明,本文提出的LSTM模型在板裂纹损伤检测中具有更高的损伤定位精度和更好的适用性.

关键词: 损伤检测, 长短期记忆神经网络, 板裂纹, 噪声

Abstract:

Aimed at the problem of intelligent classification of crack damage in different positions of the plate, a method for plate crack damage detection based on long short-term memory (LSTM) neural network is proposed. The Abaqus secondary development is used to build the plate crack damage model and calculate the acceleration response of the plate under Gaussian white noise excitation. The data set is generated by data augmentation, and the influence of noise on damage detection is considered. An intelligent crack detection model based on LSTM is established, which directly takes the acceleration response of the plate as the input and does not require additional damage feature extraction. With the goal of minimizing prediction error, the hyperparameter of the model is selected and the model configuration is optimized. The comparison of the multi-layer perceptron model and the multi-layer perceptron model based on wavelet packet transform shows that the LSTM model proposed in this paper has a higher damage location accuracy and a better applicability in plate crack detection.

Key words: damage detection, long short-term memory (LSTM) neural network, plate crack, noise

中图分类号:

U663.6
TP183

张松林, 马栋梁, 王德禹. 基于长短期记忆神经网络的板裂纹损伤检测方法[J]. 上海交通大学学报, 2021, 55(5): 527-535.

ZHANG Songlin, MA Dongliang, WANG Deyu. Method for Plate Crack Damage Detection Based on Long Short-Term Memory Neural Network[J]. Journal of Shanghai Jiao Tong University, 2021, 55(5): 527-535.

图/表 16

图1

图2

表1

表2

图3

图4

图5

表3

图6

图7

图8

图9

图10

图11

图12

图13

参考文献 13

[1]	张敬芬, 赵德有. 工程结构裂纹损伤振动诊断的发展现状和展望[J]. 振动与冲击, 2002, 21(4):22-26.
	ZHANG Jingfen, ZHAO Deyou. Summary review of vibration-based crack diagnosis technique for engingeering structures[J]. Journal of Vibration and Shock, 2002, 21(4):22-26.
[2]	贾连徽, 任慧龙, 孙树政, 等. 船体结构应力监测点的选取方法研究[J]. 船舶力学, 2013, 17(4):389-397.
	JIA Lianhui, REN Huilong, SUN Shuzheng, et al. Research on ship structural stress monitoring points selection method[J]. Journal of Ship Mechanics, 2013, 17(4):389-397.
[3]	YANG J, HWANG C N, YANG B L. Crack identification in beams and plates by discrete wavelet transform method[J]. Journal of Ship Mechanics, 2008, 12(3):464-472.
[4]	MEHRJOO M, KHAJI N, MOHARRAMI H, et al. Damage detection of truss bridge joints using Artificial Neural Networks[J]. Expert Systems with Applications, 2008, 35(3):1122-1131. doi: 10.1016/j.eswa.2007.08.008 URL
[5]	GREFF K, SRIVASTAVA R K, KOUTNÍK J, et al. LSTM: A search space odyssey[J]. IEEE Transactions on Neural Networks and Learning Systems, 2017, 28(10):2222-2232. doi: 10.1109/TNNLS.2016.2582924 URL
[6]	张建付, 宋雨, 李刚, 等. 基于长短时记忆神经网络的风电机组滚动轴承故障诊断方法[J]. 计算机测量与控制, 2017, 25(1):16-19.
	ZHANG Jianfu, SONG Yu, LI Gang, et al. A method of fault diagnosis for rolling bearing of wind turbines based on long short-term memory neural network[J]. Computer Measurement & Control, 2017, 25(1):16-19.
[7]	ZOU P, HOU B C, LEI J, et al. Bearing fault diagnosis method based on EEMD and LSTM[J]. International Journal of Computers Communications & Control, 2020, 15(1):1-14.
[8]	陈伟, 陈锦雄, 江永全, 等. 基于RS-LSTM的滚动轴承故障识别[J]. 中国科技论文, 2018, 13(10):1134-1141.
	CHEN Wei, CHEN Jinxiong, JIANG Yongquan, et al. Fault identification of rolling bearing based on RS-LSTM[J]. China Sciencepaper, 2018, 13(10):1134-1141.
[9]	YU L, QU J L, GAO F, et al. A novel hierarchical algorithm for bearing fault diagnosis based on stacked LSTM[J]. Shock and Vibration, 2019, 2019:1-10.
[10]	LUO P, HU Y. Research on rolling bearing fault Identification method based on LSTM neural network[C]//2018 International Conference on Mechanical Engineering, Materials Science and Civil Engineering, Xiamen, China: nstitute of Physics Publishing, 2019, 542:012048.
[11]	SRIVASTAVA N, HINTON G E, KRIZHEVSKY A, et al. Dropout: A simple way to prevent neural networks from overfitting[J]. Journal of Machine Learning Research, 2014, 15(1):1929-1958.
[12]	LIN Y Z, NIE Z H, MA H W. Structural damage detection with automatic feature-extraction through deep learning[J]. Computer-Aided Civil and Infrastructure Engineering, 2017, 32(12):1025-1046. doi: 10.1111/mice.2017.32.issue-12 URL
[13]	李世玲, 李治, 李合生. 基于小波包能量特征的滚动轴承故障监测方法[J]. 系统仿真学报, 2003, 15(1):76-80.
	LI Shiling, LI Zhi, LI Hesheng. The method of roller bearing fault monitoring based on wavelet packet engergy feature[J]. Journal of System Simulation, 2003, 15(1):76-80.

项目	参数
密度/(t·mm^-3)	7.85×10^-9
弹性模量/MPa	2.1×10⁵
泊松比	0.3
α/s^-1	7.2
β/s	2×10^-7

模态阶数	频率/Hz
模态阶数	简支	固支
1	38.33	81.35
2	49.79	90.54
3	68.94	107.50
4	95.84	133.21
5	130.60	167.99
6	142.98	211.80
7	154.35	222.11
8	173.34	231.92

参数	数值
时期	40
批处理大小	64
输入特征维度	11
隐藏层维度	12
LSTM层数	3
舍弃概率	0.5

[1]	范宏, 魏心武, 贾庆山, 罗佳怡. 能源-交通融合下电-气-热多能系统协同优化调度方法[J]. 上海交通大学学报, 2026, 60(2): 211-223.
[2]	. 基于CEEMDAN 和 GRU的停车位预测[J]. J Shanghai Jiaotong Univ Sci, 2025, 30(5): 962-975.
[3]	曹文平, 王尧, 张悦, 罗魁, 胡存刚, 芮涛. 计及采样噪声补偿的并网逆变器三矢量无模型预测电流控制方法[J]. 上海交通大学学报, 2025, 59(10): 1523-1532.
[4]	刘俊城, 谭勇, 张生杰. 地铁车站深基坑开挖变形智能多步预测方法[J]. 上海交通大学学报, 2024, 58(7): 1108-1117.
[5]	李芬, 孙凌, 王亚维, 屈爱芳, 梅念, 赵晋斌. 基于CEEMDAN-GSA-LSTM和SVR的光伏功率短期区间预测[J]. 上海交通大学学报, 2024, 58(6): 806-818.
[6]	姜瑞涛1，杨星2，邓又铭2，冷用斌1,2. 基于利用快速搜索和密度峰值发现进行聚类的束流位置监测器诊断方法[J]. J Shanghai Jiaotong Univ Sci, 2024, 29(3): 457-462.
[7]	于生，李向舜. 识别稳态与瞬态[J]. J Shanghai Jiaotong Univ Sci, 2024, 29(2): 261-270.
[8]	韩允, 赵新芳, 马永犇. 航空声纳浮标自噪声分析[J]. 海洋工程装备与技术, 2024, 11(1): 77-81.
[9]	尚凡成, 李传庆, 詹可, 朱仁传. 改进LSTM神经网络在极短期波浪时序预报中的应用[J]. 上海交通大学学报, 2023, 57(6): 659-665.
[10]	曹现刚1, 2，雷卓1，李彦川1，张梦园1，段欣宇1. 基于Self-Attention-LSTM神经网络的设备剩余寿命预测方法[J]. J Shanghai Jiaotong Univ Sci, 2023, 28(5): 652-664.
[11]	李婕, 司伟建, 刘佳宁. 非均匀噪声下的嵌套阵列DOA估计[J]. 空天防御, 2023, 6(4): 86-92.
[12]	刘新福, 程辉, 李晖, 李博, 高鹏, 张守森, 刘碧龙. 海上平台组块风道噪声特性及降噪措施分析[J]. 海洋工程装备与技术, 2023, 10(2): 51-56.
[13]	李苗苗, 司伟建, 颜卫忠. 脉冲噪声环境下的二维DOA估计[J]. 空天防御, 2022, 5(2): 94-100.
[14]	孙翀, 田甜, 竺晓程, 杜朝辉. 风力机翼型非定常流场POD和EPOD分析[J]. 上海交通大学学报, 2022, 56(1): 45-52.
[15]	谭嘉, 李知艺, 杨欢, 赵荣祥, 鞠平. 基于分布式优化思想的配电网用电负荷多层协同预测方法[J]. 上海交通大学学报, 2021, 55(12): 1544-1553.

基于长短期记忆神经网络的板裂纹损伤检测方法

Method for Plate Crack Damage Detection Based on Long Short-Term Memory Neural Network

RichHTML

PDF (PC)

可视化

被引次数

摘要/Abstract

引用本文

使用本文

图/表 16

参考文献 13

相关文章 15

编辑推荐

Metrics

本文评价