用于腐蚀监测的电化学噪声信号识别方法

doi:10.11902/1005.4537.2024.336

中国腐蚀与防护学报

2025, Vol. 45

Issue (5): 1433-1440 CSTR: 32134.14.1005.4537.2024.336 DOI: 10.11902/1005.4537.2024.336

研究报告

本期目录 | 过刊浏览

用于腐蚀监测的电化学噪声信号识别方法

申志远¹^,², 单广斌¹^,², 陈闽东¹^,²(

), 刘媛双¹^,²

1 化学品安全全国重点实验室青岛 266104
2 中石化安全工程研究院有限公司青岛 266104

An Electrochemical Noise Signal Recognition Method for Corrosion Monitoring

SHEN Zhiyuan¹^,², SHAN Guangbin¹^,², CHEN Mindong¹^,²(

), LIU Yuanshuang¹^,²

1 State Key Laboratory of Chemical Safety, Qingdao 266104, China
2 SINOPEC Research Institute of Safety Engineering Co., Ltd., Qingdao 266104, China

引用本文:

申志远, 单广斌, 陈闽东, 刘媛双. 用于腐蚀监测的电化学噪声信号识别方法[J]. 中国腐蚀与防护学报, 2025, 45(5): 1433-1440.
Zhiyuan SHEN, Guangbin SHAN, Mindong CHEN, Yuanshuang LIU. An Electrochemical Noise Signal Recognition Method for Corrosion Monitoring[J]. Journal of Chinese Society for Corrosion and protection, 2025, 45(5): 1433-1440.

全文: PDF(4657 KB) HTML

摘要:

为实现对实时采集的电化学噪声信号进行自动且准确的分析，本文提出一种用于腐蚀监测的电化学噪声信号识别网络。使用最大池化运算进行信号平滑，同时保留信号的细节特征和趋势特征，并以网络模块设计的方式实现端到端训练；在残差结构和空间金字塔池化结构的基础上设计了特征提取模块，以强化网络对关键特征的表征能力；在实验获得的电化学噪声信号上采用5折交叉验证的方式对模型训练与测试。实验结果表明，所提出的模型总体准确率和F1分数分别达到0.9463和0.9282；利用神经网络直接建模可以实现对电化学噪声信号的准确识别。

关键词 ：电化学噪声, 神经网络, 深度学习, 模式识别

Abstract：

An electrochemical noise signal recognition network for corrosion monitoring was proposed in order to realize the automatic and accurate analysis of electrochemical noise signals collected in real time. The maximum pooling operation method was adopted to smooth the signal while the details and trend features of the signal are preserved, and the end-to-end training is realized by using a network module design. Based on residual structures and spatial pyramid pooling structures, a feature extraction module was designed to enhance the network's ability to characterize the key features. The model was trained and tested via 5-fold cross-validation based on the experimentally acquired electrochemical noise signals. The results show that the proposed model achieved an overall accuracy and F1 score of 0.9463 and 0.9282, respectively, demonstrating that neural networks can be used to accurately identify electrochemical noise signals.

Key words： electrochemical noise neural network deep learning pattern recognition

收稿日期: 2024-10-12 32134.14.1005.4537.2024.336

ZTFLH:

TG172

基金资助:国家重点研发计划(2022YFC3004502)

通讯作者: 陈闽东，E-mail：chenmd.qday@sinopec.com，研究方向为腐蚀与防护

Corresponding author: CHEN Mindong, E-mail: chenmd.qday@sinopec.com

作者简介: 申志远，男，1997年生，硕士，助理工程师

	服务

	把本文推荐给朋友
	加入引用管理器
	E-mail Alert
	RSS
	作者相关文章
	申志远
	单广斌
	陈闽东
	刘媛双
44.8K

链接本文:

https://www.jcscp.org/CN/10.11902/1005.4537.2024.336 或 https://www.jcscp.org/CN/Y2025/V45/I5/1433

图1 不同腐蚀阶段下电位与电流信号对比

图2 304不锈钢在3.5%NaCl溶液中电化学噪声实验前后的SEM表面形貌对比

图3 各种信号平滑方式对比

图4 最大池化运算前后电化学噪声信号对比

图5 集成信号平滑空间金字塔池化(SPPSS)与残差空间金字塔池化(RSPP)的网络架构

图6 SPPSS模块计算过程示意图

图7 RSPP模块计算过程示意图

图8 训练和测试过程中的损失值下降

图9 SPPSS-RSPP架构的五折交叉验证ROC性能

表1 不同随机种子下网络的准确率/%

图10 网络的F1分数和准确率对比

图11 各网络检测结果的混淆矩阵

[1]	Zhang Z, Li X F, Zhao Z Y, et al. In-situ monitoring of pitting corrosion of Q235 carbon steel by electrochemical noise: Wavelet and recurrence quantification analysis [J]. J. Electroanal. Chem., 2020, 879: 114776
[2]	Zhang H F, Wu Z Q, Chen Y, et al. Real-time monitoring of the corrosion behaviour of the 304SS in HCl solution using BPNN with joint image recognition and electrochemical noise [J]. Corros. Sci., 2024, 228: 111779
[3]	Jiao Y B, Zhang D P, Yu J, et al. Early identification of stress corrosion cracking of P110 low alloy steel in downhole fluid by electrochemical noise measurement [J]. Corros. Eng. Sci. Technol., 2021, 56: 230
[4]	Shi P F, Hu L Y, Duan G Q, et al. Research and prospect of corrosion monitoring techniques [J]. Total. Corros. Control, 2023, 37(10): 23
[4]	石鹏飞, 胡凌越, 段国庆等. 腐蚀监测技术研究现状与展望 [J]. 全面腐蚀控制, 2023, 37(10): 23
[5]	Han L, Zhang Y L, Liu X H, et al. The application of electrochemical noise technique on refinery technique on refinery [J]. Safety Health Environ., 2015, 15(5): 50
[5]	韩磊, 张艳玲, 刘小辉等. 电化学噪声技术在炼油厂腐蚀监测中的应用 [J]. 安全、健康和环境, 2015, 15(5): 50
[6]	Zhou M X, Wu J, Fan Z B, et al. Current situation and prospect of on-line monitoring technology for atmospheric corrosion testing of metallic materials [J]. J. Chin. Soc. Corros. Prot., 2023, 43: 38
[6]	周梦鑫, 吴军, 樊志彬等. 大气腐蚀在线监测技术研究现状与展望 [J]. 中国腐蚀与防护学报, 2023, 43: 38 doi: 10.11902/1005.4537.2022.027
[7]	Sun Z J, Xue L, Xu Y M, et al. Overview of deep learning [J]. Appl. Res. Comput., 2012, 29: 2806
[7]	孙志军, 薛磊, 许阳明等. 深度学习研究综述 [J]. 计算机应用研究, 2012, 29: 2806
[8]	Hu Q F, Zhang T, Chen S H, et al. An instantaneous corrosion monitoring technique based on combining modified electrochemical noise and artificial neural network for determination of corrosion type and 2014 aluminium alloy corrosion rate in NaCl and Ce(NO₃)₃ solutions [J]. Int. J. Electrochem. Sci., 2022, 17: 220213
[9]	Nazarnezhad-Bajestani M, Neshati J, Siadati M H. Determination of SS321 pitting stage in FeCl₃ solution based on electrochemical noise measurement data using artificial neural network [J]. J. Electroanal. Chem., 2019, 845: 31 doi: 10.1016/j.jelechem.2019.05.036
[10]	Calabrese L, Galeano M, Proverbio E, et al. Topological neural network of combined AE and EN signals for assessment of SCC damage [J]. Nondestruct. Test. Eval., 2020, 35(1): 98
[11]	Liu W. Asymmetric surface configuration for electrochemical noise measurement on stainless steel [J]. J. Chin. Soc. Corros. Prot., 2023, 43: 1151
[11]	刘微. 测量不锈钢电化学噪声的非对称表面方法 [J]. 中国腐蚀与防护学报, 2023, 43: 1151
[12]	An P L, Liang P, Ren J M, et al. Characteristics on electrochemical noise of pitting corrosion for high nitrogen austenitic stainless steels [J]. J. Chin. Soc. Corros. Prot., 2018, 38: 26
[12]	安朋亮, 梁平, 任建民等. 高氮奥氏体不锈钢点蚀行为的电化学噪声特征 [J]. 中国腐蚀与防护学报, 2018, 38: 26 doi: 10.11902/1005.4537.2016.242
[13]	Hu H H, Wang G, Deng P C, et al. Electrochemical noise characteristics of 304 stainless steel pitting based on MATLAB [J]. J. Guangdong Ocean Univ., 2019, 39(4): 89
[13]	胡欢欢, 王贵, 邓培昌等. 基于MATLAB的304不锈钢点蚀行为电化学噪声特征 [J]. 广东海洋大学学报, 2019, 39(4): 89
[14]	Li Z G, Peng D H, Pu J W, et al. Pitting corrosion of stainless steel in chloride-containing potassium sulfate solution based on electrochemical noise technique [J]. Mater. Prot., 2018, 51(9): 122
[14]	李志刚, 彭东辉, 浦静雯. 基于电化学噪声研究不锈钢在含氯硫酸钾溶液中的点蚀特征 [J]. 材料保护, 2018, 51(9): 122
[15]	Deng J H, Wang G, Hu J Z, et al. Pitting behavior of stainless steel in simulated marine atmosphere based on electrochemical noise [J]. J. Electrochem., 2020, 26: 298
[15]	邓俊豪, 王贵, 胡杰珍等. 基于电化学噪声研究模拟海洋大气环境下304不锈钢的点蚀行为 [J]. 电化学, 2020, 26: 298 doi: 10.13208/j.electrochem.190415
[16]	Li Z G, Peng D H, Pu J W, et al. Application of electrochemical noise technology to the study of pitting behavior of 2205 steel [J]. Corros. Prot., 2018, 39: 11
[16]	李志刚, 彭东辉, 浦静雯等. 电化学噪声技术在2205钢点蚀行为研究中的应用 [J]. 腐蚀与防护, 2018, 39: 11
[17]	Deng J H. Research and application of electrochemical noise technology in atmospheric pitting monitoring of stainless steel [D]. Zhanjiang: Guangdong Ocean University, 2020
[17]	邓俊豪. 电化学噪声技术在不锈钢大气点蚀监测中的研究与应用 [D]. 湛江: 广东海洋大学, 2020
[18]	Hu H H, Deng P C, Hu J Z, et al. Monitoring method of carbon steel atmosphere corrosion behavior based on electrochemical noise technology [J]. Eq. Envir. Eng., 2017, 14(9): 68
[18]	胡欢欢, 邓培昌, 胡杰珍等. 基于电化学噪声技术的碳钢大气腐蚀行为监测方法 [J]. 装备环境工程, 2017, 14(9): 68
[19]	Wang F P, Jiang X Y, Zhang H. The relationship between the electrochemical noise spectrogram and the surface state of AZ91D magnesium alloy in sodium chloride solution [J]. J. Liaoning Norm. Univ. (Nat. Sci. Ed.), 2015, 38: 497
[19]	王凤平, 蒋新瑜, 张航. AZ91D镁合金在NaCl溶液中电化学噪声谱与表面状态关系 [J]. 辽宁师范大学学报(自然科学版), 2015, 38: 497
[20]	Zhou F Y, Jin L P, Dong J. Review of convolutional neural network [J]. Chin. J. Comput., 2017, 40: 1229
[20]	周飞燕, 金林鹏, 董军. 卷积神经网络研究综述 [J]. 计算机学报, 2017, 40: 1229
[21]	Liu B Y. Research on signal peak preservation smoothing algorithm based on segmented classification [D]. Nanjing: Nanjing University of Information Science & Technology, 2021
[21]	刘宝莹. 基于分段分类的信号保峰平滑算法研究 [D]. 南京: 南京信息工程大学, 2021
[22]	He K M, Zhang X Y, Ren S Q, et al. Deep residual learning for image recognition [A]. Proceedings of the 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) [C]. Vegas, 2016: 770
[23]	Huang W B, Chen R W, Yuan T T. Compression of UAV object detection model based on improved YOLOv3-SPP [J]. Comput. Eng. Appl., 2021, 57(21): 165 doi: 10.3778/j.issn.1002-8331.2007-0230
[23]	黄文斌, 陈仁文, 袁婷婷. 改进YOLOv3-SPP的无人机目标检测模型压缩方案 [J]. 计算机工程与应用, 2021, 57(21): 165 doi: 10.3778/j.issn.1002-8331.2007-0230
[24]	Fu H T, Wang P, Li X Y, et al. Lightweight network model for moving object recognition [J]. J. Xi'an Jiaotong Univ., 2021, 55(7): 124
[24]	符惠桐, 王鹏, 李晓艳等. 面向移动目标识别的轻量化网络模型 [J]. 西安交通大学学报, 2021, 55(7): 124
[25]	Zhou Z H. Machine Learning [M]. Beijing: Tsinghua University Press, 2016: 28
[25]	周志华. 机器学习 [M]. 北京: 清华大学出版社, 2016: 28
[26]	Carrington A M, Manuel D G, Fieguth P W, et al. Deep ROC analysis and AUC as balanced average accuracy, for improved classifier selection, audit and explanation [J]. IEEE Trans. Pattern Anal. Mach. Intell., 2023, 45: 329

[1]	周涛涛, 刘迎正, 郑文培, 姜恒良, 刘海鹏, 夏刚. 基于物理信息神经网络的油气管道内腐蚀预测方法[J]. 中国腐蚀与防护学报, 2025, 45(5): 1320-1330.
[2]	李婕, 孟凡帝, 孙学思, 李佳妮, 陈思涵, 李则蓝, 迟剑宁, 亓海霞, 王福会, 刘莉. 基于多尺度图像特征融合的有机涂层寿命预测研究[J]. 中国腐蚀与防护学报, 2025, 45(5): 1205-1218.
[3]	陈潜, 黄伟, 张昌会, 管奥成, 张宸, 叶晓芃. 基于物理引导的集输管道内腐蚀速率预测及可解释性分析[J]. 中国腐蚀与防护学报, 2025, 45(3): 720-730.
[4]	王亚东, 马荣耀, 万晔, 董俊华. 静水压力对吉帕级海工钢母材及焊材腐蚀行为的影响[J]. 中国腐蚀与防护学报, 2025, 45(3): 653-663.
[5]	刘微. 测量不锈钢电化学噪声的非对称表面方法[J]. 中国腐蚀与防护学报, 2023, 43(5): 1151-1158.
[6]	王明好, 王欢, 刘叡, 孟凡帝, 刘莉, 王福会. 基于深度学习方法的N5/NiCrAlY涂层图像识别的研究[J]. 中国腐蚀与防护学报, 2022, 42(4): 583-589.
[7]	李鸿瑾, 王歧山, 廖子涵, 孙祥锐, 孙晖, 张新阳, 陈旭. X70钢及其焊缝在含Cl^-高pH值溶液中电化学噪声行为研究[J]. 中国腐蚀与防护学报, 2022, 42(1): 60-66.
[8]	曹京宜, 方志刚, 李亮, 冯亚菲, 王兴奇, 寿海明, 杨延格, 褚广哲, 殷文昌. 国产镀锌钢在不同水环境中的腐蚀行为：I淡水和盐水[J]. 中国腐蚀与防护学报, 2021, 41(2): 169-177.
[9]	张震, 吴欣强, 谭季波. 电化学噪声原位监测应力腐蚀开裂的研究现状与进展[J]. 中国腐蚀与防护学报, 2020, 40(3): 223-229.
[10]	安朋亮, 梁平, 任建民, 史艳华, 刘峰, 陈思瑶. 高氮奥氏体不锈钢点蚀行为的电化学噪声特征[J]. 中国腐蚀与防护学报, 2018, 38(1): 26-32.
[11]	郭宝会, 邱友绪, 李海龙. 人工神经网络在钛合金表面Ni-SiC复合电镀工艺中的应用[J]. 中国腐蚀与防护学报, 2017, 37(4): 389-394.
[12]	陈洁净,鞠虹,孙灿,李霞,刘雲飞. 电化学测试技术在垢下腐蚀中的应用[J]. 中国腐蚀与防护学报, 2017, 37(3): 207-215.
[13]	刘静,李晓禄,朱崇伟,张涛,曾冠鑫,孟国哲,邵亚薇. 利用人工神经网络技术预测气田环境下316L不锈钢临界点蚀温度[J]. 中国腐蚀与防护学报, 2016, 36(3): 205-211.
[14]	邓志安,李姝仪,李晓坤,王珊,王晓军. 基于模糊神经网络的海洋管线腐蚀速率预测新方法[J]. 中国腐蚀与防护学报, 2015, 35(6): 571-576.
[15]	石维, 董泽华, 郭兴蓬. 基于Hilbert-Huang变换的电化学噪声解析及其应用[J]. 中国腐蚀与防护学报, 2014, 34(2): 138-146.

Viewed

Full text

Abstract

Cited

Shared

Discussed