上海理工大学 管理学院,上海 200093
李文赞,男,2002年生,山东枣庄人,在读硕士研究生;主要研究方向为故障诊断;E-mail:leewzzz@163.com。
刘勤明(通信作者),男,1984年生,山东日照人,博士,教授,博士研究生导师;主要研究方向为维护调度、人工智能等;E-mail:lqm0531@163.com。
收稿:2025-06-16,
纸质出版:2026-02-15
移动端阅览
李文赞,刘勤明,叶春明,等. 基于深度学习与SMOTE数据增强的旋转机械故障诊断研究[J]. 机械强度,2026,48(2):1-11.
LI Wenzan,LIU Qinming,YE Chunming,et al. Research on fault diagnosis of rotating machinery based on deep learning and SMOTE data enhancement[J]. Journal of Mechanical Strength,2026,48(2):1-11.
李文赞,刘勤明,叶春明,等. 基于深度学习与SMOTE数据增强的旋转机械故障诊断研究[J]. 机械强度,2026,48(2):1-11. DOI: 10.16579/j.issn.1001.9669.2026.02.001.
LI Wenzan,LIU Qinming,YE Chunming,et al. Research on fault diagnosis of rotating machinery based on deep learning and SMOTE data enhancement[J]. Journal of Mechanical Strength,2026,48(2):1-11. DOI: 10.16579/j.issn.1001.9669.2026.02.001.
目的
2
随着工业设备自动化与智能化的发展,旋转机械的故障诊断已成为保障设备稳定运行的关键环节。传统故障诊断方法依赖人工经验和简单信号处理技术,难以应对复杂工况和多样化的故障类型。因此,提出一种基于格拉姆角场(Gram Angular Field
GAF)图像编码和增强型门控波形网络(Gated WaveNet
GWaveNet)架构的新型故障诊断方法。
方法
2
首先,将原始一维轴承振动信号通过GAF编码转换为二维图像,充分保留时序依赖性和动态信号变化特征。在这一过程中,为解决数据不平衡问题,采用窗口切分、噪声增强、合成少数类过采样技术(Synthetic Minority Over-sampling Technique
SMOTE)等扩增样本多样性,增强了模型的训练效果并提高了鲁棒性。其次,构建了一种增强型GWaveNet架构,结合卷积层、残差连接、多尺度特征提取机制,增强了对故障模式的辨识能力。最后,基于美国凯斯西储大学(Case Western Reserve University
CWRU)公共数据集进行试验验证。试验结果表明,所提方法能够在健康、内圈故障、外圈故障3种工况下实现高准确率的故障诊断,并在复杂工况下展示出较强的泛化能力。
结果
2
结果表明,所提模型在旋转机械故障诊断中具有较高的准确性和鲁棒性,能够有效应对故障类型的多样性和实际工业环境中的不确定性。
Objective
2
With the advancement of automation and intelligence in industrial equipment
fault diagnosis of rotating machinery has become a critical component in ensuring the stable operation of equipment. Traditional diagnostic approaches
which often depend on expert knowledge and basic signal processing techniques
struggle to manage complex operating conditions and varied fault types. To address these challenges
a novel fault diagnosis framework was introduced
leveraging Gram angular field (GAF) image encoding and an enhanced gated WaveNet (GWaveNet) architecture.
Methods
2
Firstly
the original one-dimensional bearing vibration signals were transformed into two-dimensional images via GAF encoding
effectively preserving temporal dependencies and dynamic signal variations. In this process
to mitigate issues related to data imbalance
techniques such as window segmentation
noise augmentation
and synthetic minority over-sampling technique (SMOTE) were employed
thereby increasing sample diversity and improving model's robustness and training efficacy. Secondly
the enhanced GWaveNet architecture integrated convolutional layers
residual connections
and multi-scale feature extraction mechanism
which collectively strengthen the network's capacity to recognize different fault patterns. Finally
test validation on the Case Western Reserve University (CWRU) public dataset was performed. The test results demonstrated that the proposed method can achieve high diagnostic accuracy under three operating conditions—normal
inner race fault
and outer race fault
while exhibiting strong generalization performance in complex operating conditions.
Results
2
The results demonstrate that the proposed model exhibits high accuracy and robustness in fault diagnosis of rotating machinery
and can effectively address a wide range of fault types and uncertainties inherent in real industrial environments.
苏莫涵 . 基于多源传感器的滚动轴承故障诊断 [J]. 信息记录材料 , 2024 , 25 ( 12 ): 73 - 76 .
SU Mohan . Fault diagnosis of rolling bearings based on multi-source sensors [J]. Information Recording Materials , 2024 , 25 ( 12 ): 73 - 76 . (In Chinese)
王玉敏 , 吴庭苇 , 苏荔 , 等 . 基于卷积神经网络的进动故障诊断方法 [C]// 中国核科学技术进展报告(第八卷) 中国核学会2023年学术年会论文集 第10册 核仪器 核石墨及碳材料测试与应用 数字化与系统工程 核质量保证 核电运行财 核心理研究与培训 . 北京 : 中国核学会 , 2023 : 80 - 85 .
WANG Yumin , WU Tingwei , SU Li , et al . A precession fault diagnosis method based on convolutional neural network [C]// Progress Report on Nuclear Science and Technology in China (Volume 8), Proceedings of the 2023 Annual Academic Conference of the Chinese Nuclear Society, Volume 10: Nuclear Instruments, Nuclear Graphite and Carbon Materials Testing and Applications, Digitalization and Systems Engineering, Nuclear Quality Assurance, Nuclear Power Plant Operations, Nuclear Psychology Research and Training . Beijing : Chinese Nuclear Society , 2023 : 80 - 85 . (In Chinese)
陈钱 , 陈康康 , 董兴建 , 等 . 一种面向机械设备故障诊断的可解释卷积神经网络 [J]. 机械工程学报 , 2024 , 60 ( 12 ): 65 - 76 .
CHEN Qian , CHEN Kangkang , DONG Xingjian , et al . Interpretable convolutional neural network for mechanical equipment fault diag-nosis [J]. Journal of Mechanical Engineering , 2024 , 60 ( 12 ): 65 - 76 . (In Chinese)
李梓煊 , 刘宁 , 张晓蕊 , 等 . 基于改进卷积神经网络的旋转机械故障诊断研究 [J]. 辽宁科技学院学报 , 2025 , 27 ( 1 ): 24 - 29 .
LI Zixuan , LIU Ning , ZHANG Xiaorui , et al . Study on fault diagnosis of rotating machinery based on improved convolutional neural network [J]. Journal of Liaoning Institute of Science and Technology , 2025 , 27 ( 1 ): 24 - 29 . (In Chinese)
姜烨飞 , 王华 , 潘裕斌 , 等 . 基于GADF-MDSC的特大型轴承深度迁移故障诊断方法 [J]. 振动与冲击 , 2024 , 43 ( 19 ): 10 - 18 .
JIANG Yefei , WANG Hua , PAN Yubin , et al . Deep transfer fault diagnosis method for extra-large bearings based on GADF-MDSC [J]. Journal of Vibration and Shock , 2024 , 43 ( 19 ): 10 - 18 . (In Chinese)
翟道宇 , 孙燕楠 . 基于卷积神经网络和格拉姆角差场的四象限脉冲整流器故障诊断方法 [J]. 电气技术 , 2025 , 26 ( 1 ): 23 - 32 .
ZHAI Daoyu , SUN Yannan . Fault diagnosis method for four-quadrant pulse rectifiers based on convolutional neural network and Gramian angular difference field [J]. Electrical Engineering , 2025 , 26 ( 1 ): 23 - 32 . (In Chinese)
刘迪洋 , 张清华 , 胡勤 . 基于深度学习的轴承故障小样本扩容及智能诊断 [J]. 机床与液压 , 2025 , 53 ( 4 ): 10 - 18 .
LIU Diyang , ZHANG Qinghua , HU Qin . Deep learning-based small sample expansion and intelligent diagnosis of bearing faults [J]. Machine Tool & Hydraulics , 2025 , 53 ( 4 ): 10 - 18 . (In Chinese)
赵国威 , 曾静 . 基于GAF和SE-ResNeXt的滚动轴承故障诊断方法 [J]. 机械设计与制造工程 , 2025 , 54 ( 2 ): 63 - 67 .
ZHAO Guowei , ZENG Jing . Fault diagnosis of the rolling bearing based on GAF and SE-ResNeXt [J]. Machine Design and Manufac-turing Engineering , 2025 , 54 ( 2 ): 63 - 67 . (In Chinese)
古莹奎 , 吴宽 , 李成 . 基于格拉姆角场和迁移深度残差神经网络的滚动轴承故障诊断 [J]. 振动与冲击 , 2022 , 41 ( 21 ): 228 - 237 .
GU Yingkui , WU Kuan , LI Cheng . Rolling bearing fault diagnosis based on Gram angle field and transfer deep residual neural network [J]. Journal of Vibration and Shock , 2022 , 41 ( 21 ): 228 - 237 . (In Chinese)
屈绍宇 , 李沅 , 卢研宏 . 基于格拉姆角场边缘实现的轴承故障诊断系统 [J]. 仪表技术与传感器 , 2024 ( 11 ): 66 - 72 .
QU Shaoyu , LI Yuan , LU Yanhong . Bearing fault diagnosis system based on edge implementation of Gramian angle field [J]. Instrument Technique and Sensor , 2024 ( 11 ): 66 - 72 . (In Chinese)
苏斌 , 侯思祖 , 郭威 . 基于图像融合和双通道卷积神经网络的配电网故障选线方法研究 [J]. 电子测量与仪器学报 , 2024 , 38 ( 9 ): 54 - 66 .
SU Bin , HOU Sizu , GUO Wei . Research on distribution network fault line selection method based on image fusion and dual-channel convolutional neural network [J]. Journal of Electronic Measurement and Instrumentation , 2024 , 38 ( 9 ): 54 - 66 . (In Chinese)
李佰霖 , 陈昱锐 , 褚凡武 , 等 . 基于混合采样和CAWOA-BP的变压器故障诊断 [J]. 水电能源科学 , 2024 , 42 ( 3 ): 216 - 220 .
LI Bailin , CHEN Yurui , CHU Fanwu , et al . Transformer fault diagnosis based on hybrid sampling and CAWOA-BP [J]. Water Resources and Power , 2024 , 42 ( 3 ): 216 - 220 . (In Chinese)
WEN T , LI J K , FEI R , et al . Dual-input robust diagnostics for railway point machines via audio signals [J]. Network:Computation in Neural Systems , 2025 , 36 ( 3 ): 1163 - 1184 .
LIU D , ZHONG S S , LIN L , et al . Feature-level SMOTE: augmenting fault samples in learnable feature space for imbalanced fault diagnosis of gas turbines [J]. Expert Systems with Applications , 2024 , 238 : 122023 .
张昊 , 王海茹 , 马继东 . 基于数据增强的可解释旋转机械故障诊断 [J]. 电子测量技术 , 2025 , 48 ( 8 ): 105 - 115 .
ZHANG Hao , WANG Hairu , MA Jidong . Inter-pretable rotating machinery fault diagnosis based on data enhancement [J]. Electro-nic Measurement Technology , 2025 , 48 ( 8 ): 105 - 115 . (In Chinese)
吴轲 , 吴军 , 舒启明 , 等 . 基于深度学习的旋转机械小样本故障诊断方法研究综述 [J]. 中国舰船研究 , 2025 , 20 ( 2 ): 3 - 19 .
WU Ke , WU Jun , SHU Qiming , et al . A review of deep learning-based few sample fault diagnosis method for rotating machinery [J]. Chinese Journal of Ship Research , 2025 , 20 ( 2 ): 3 - 19 . (In Chinese)
TANG T , WANG J W , YANG T Y , et al . An improved prototypical network with L2 prototype correction for few-shot cross-domain fault diagnosis [J]. Measurement , 2023 , 217 : 113065 .
李靖宇 , 董增寿 , 康琳 , 等 . 基于WGAN-div和DLA34模型的轴承故障智能诊断新方法 [J]. 机床与液压 , 2024 , 52 ( 19 ): 225 - 236 .
LI Jingyu , DONG Zengshou , KANG Lin , et al . A new method of bearing fault intelligent diagnosis based on WGAN-div and DLA34 models [J]. Machine Tool & Hydraulics , 2024 , 52 ( 19 ): 225 - 236 . (In Chinese)
RACHEAL S , LIU Y K , AYODEJI A . Improved WaveNet for pressurized water reactor accident prediction [J]. Annals of Nuclear Energy , 2023 , 181 : 109519 .
0
浏览量
52
下载量
0
CSCD
关联资源
相关文章
相关作者
相关机构
京公网安备11010802024621