南京航空航天大学 机电学院,南京 210016
陈富星,男,2000年生,山东临沂人,硕士研究生;主要研究方向为齿轮箱故障诊断与深度学习;E-mail:chenfuxing@nuaa.edu.cn。
收稿:2024-05-15,
修回:2024-07-08,
纸质出版:2026-02-15
移动端阅览
陈富星,冷晟,高海淋,等. 注意力机制下多域特征融合的齿轮箱故障诊断方法[J]. 机械强度,2026,48(2):12-20.
CHEN Fuxing,LENG Sheng,GAO Hailin,et al. Gearbox fault diagnosis method with multi-domain feature fusion based on attention mechanism[J]. Journal of Mechanical Strength,2026,48(2):12-20.
陈富星,冷晟,高海淋,等. 注意力机制下多域特征融合的齿轮箱故障诊断方法[J]. 机械强度,2026,48(2):12-20. DOI: 10.16579/j.issn.1001.9669.2026.02.002.
CHEN Fuxing,LENG Sheng,GAO Hailin,et al. Gearbox fault diagnosis method with multi-domain feature fusion based on attention mechanism[J]. Journal of Mechanical Strength,2026,48(2):12-20. DOI: 10.16579/j.issn.1001.9669.2026.02.002.
目的
2
针对单一数据域模型在齿轮箱故障诊断中难以精准识别微弱故障特征的局限性,提出一种注意力机制下多域特征融合的故障诊断方法,以提升诊断准确率、稳定性及泛化能力。
方法
2
首先,从振动信号的时域、频域提取无量纲特征与频谱特征。其次,通过连续小波变换结合卷积神经网络(Convolutional Neural Network
CNN)提取时-频域深度特征。然后,引入注意力机制对多域特征进行动态加权融合,强化关键特征并弱化冗余信息。最后,通过分类器完成故障识别,基于二级齿轮箱试验数据集验证方法有效性。
结果
2
在二级齿轮箱试验数据集上的测试结果表明,所提方法诊断准确率达到99.77%,优于单一域模型,验证了其在微弱故障识别与多工况适应方面的有效性与稳定性。
Objective
2
Aiming at the limitation that single data domain models are difficult to accurately identify subtle fault features in gearbox fault diagnosis
a fault diagnosis method with multi-domain feature fusion under attention mechanism was proposed to improve diagnostic accuracy
stability and generalization ability.
Methods
2
Firstly
dimensionless features and spectral features were extracted from the time domain and frequency domain of vibration signals. Secondly
deep time-frequency domain features were extracted by combining continuous wavelet transform with convolutional neural network (CNN). Then
an attention mechanism was introduced to dynamically weight and fuse multi-domain features
strengthening key features and weakening redundant information. Finally
a classifier was used to complete fault identification
and the effectiveness of the method was verified based on a secondary gearbox test dataset.
Results
2
Test results on a secondary gearbox test dataset show that the proposed method achieves a diagnostic accuracy of 99.77%
outperforming single-domain models and verifying its effectiveness and stability in identifying weak faults and adapting to multiple operating conditions.
TONG S G , LI J J , CONG F Y , et al . Vibration separation methodology compensated by time-varying transfer function for fault diagnosis of non-hunting tooth planetary gearbox [J]. Sensors , 2022 , 22 ( 2 ): 557 .
王辉 , 徐佳文 , 严如强 . 基于多尺度注意力深度强化学习网络的行星齿轮箱智能诊断方法 [J]. 机械工程学报 , 2022 , 58 ( 11 ): 133 - 142 .
WANG Hui , XU Jiawen , YAN Ruqiang . Multi-scale attention based deep reinforcement learning for intelligent fault diagnosis of planetary gearbox [J]. Journal of Mechanical Engineering , 2022 , 58 ( 11 ): 133 - 142 . (In Chinese)
雷彪 , 惠恩明 , 关海英 , 等 . 基于双种群协调进化果蝇算法优化SVM的变速箱故障诊断 [J]. 机械强度 , 2022 , 44 ( 3 ): 753 - 757 .
LEI Biao , HUI Enming , GUAN Haiying , et al . Gearbox diagnosis based on SVM optimized by double group coevolution fruit fly optimization algorithm [J]. Journal of Mechanical Strength , 2022 , 44 ( 3 ): 753 - 757 . (In Chinese)
戚晓利 , 崔创创 , 杨艳 , 等 . 基于RTSMFE、M-KRCDA与COA-SVM的行星齿轮箱故障诊断 [J]. 振动与冲击 , 2022 , 41 ( 21 ): 109 - 120 .
QI Xiaoli , CUI Chuangchuang , YANG Yan , et al . Planetary gearbox fault diagnosis based on RTSMFE,M-KRCDA and COA-SVM [J]. Journal of Vibration and Shock , 2022 , 41 ( 21 ): 109 - 120 . (In Chinese)
张书瑶 , 王梓齐 , 刘长良 . 基于改进集成KNN回归算法的风电动机组齿轮箱状态监测 [J]. 动力工程学报 , 2023 , 43 ( 6 ): 759 - 767 .
ZHANG Shuyao , WANG Ziqi , LIU Changliang . Wind turbine gearbox condition monitoring based on improved ensemble KNN regression algorithm [J]. Journal of Chinese Society of Power Engineering , 2023 , 43 ( 6 ): 759 - 767 . (In Chinese)
IKHLEF B , RAHMOUNE C , BETTAHAR T , et al . Gearboxes fault detection under operation varying condition based on MODWPT,Ant colony optimization algorithm and Random Forest classifier [J]. Advances in Mechanical Engineering , 2021 , 13 ( 8 ): 7 - 12 .
王博 , 南新元 . 基于AO-VMD和IAO-SVM的齿轮箱故障诊断 [J]. 机械传动 , 2023 , 47 ( 5 ): 143 - 149 .
WANG Bo , NAN Xinyuan . Fault diagnosis of gearboxes based on AO-VMD and IAO-SVM [J]. Journal of Mechanical Transmission , 2023 , 47 ( 5 ): 143 - 149 . (In Chinese)
MA B , CAI W D , HAN Y M , et al . A novel probability confidence CNN model and its application in mechanical fault diagnosis [J]. IEEE Transactions on Instrumentation and Measurement , 2021 , 70 : 1 - 11 .
黄佳欢 , 余建波 . 基于形态帽积卷积自编码器模型的齿轮箱故障诊断 [J]. 机械工程学报 , 2023 , 59 ( 23 ): 132 - 145 .
HUANG Jiahuan , YU Jianbo . Gearbox fault diagnosis research based on morphological hat product convolutional autoencoder [J]. Journal of Mechanical Engineering , 2023 , 59 ( 23 ): 132 - 145 . (In Chinese)
张龙 , 徐天鹏 , 王朝兵 , 等 . 基于卷积门控循环网络的齿轮箱故障诊断 [J]. 吉林大学学报(工学版) , 2022 , 52 ( 2 ): 368 - 376 .
ZHANG Long , XU Tianpeng , WANG Chaobing , et al . Gearbox fault diagnosis baed on convolutional gated recurrent network [J]. Journal of Jilin University(Engineering and Technology Edition) , 2022 , 52 ( 2 ): 368 - 376 . (In Chinese)
戚晓利 , 程主梓 , 崔创创 , 等 . 基于JS-VME-DBN和MS-UMAP的行星齿轮箱故障诊断方法 [J]. 航空动力学报 , 2024 , 39 ( 3 ): 236 - 248 .
QI Xiaoli , CHENG Zhuzi , CUI Chuangchuang , et al . Fault diagnosis method of planetary gearbox based on JS-VME-DBN and MS-UMAP [J]. Journal of Aerospace Power , 2024 , 39 ( 3 ): 236 - 248 . (In Chinese)
杜浩飞 , 张超 , 李建军 . 基于SENet-ResNext-LSTM的风机轴承故障诊断 [J]. 机械强度 , 2023 , 45 ( 6 ): 1271 - 1279 .
DU Haofei , ZHANG Chao , LI Jianjun . Fault diagnosis of wind turbine bearing based on SENet-ResNext-LSTM [J]. Journal of Mechanical Strength , 2023 , 45 ( 6 ): 1271 - 1279 . (In Chinese)
王敏 , 邓艾东 , 马天霆 , 等 . 基于双注意力机制的MSCN-BiGRU的滚动轴承故障诊断方法 [J]. 振动与冲击 , 2024 , 43 ( 6 ): 84 - 92 .
WANG Min , DENG Aidong , MA Tianting , et al . Rolling bearing fault diagnosis method based on a multi-scale and improved gated recurrent neural network with dual attention [J]. Journal of Vibration and Shock , 2024 , 43 ( 6 ): 84 - 92 . (In Chinese)
HAN B K , LI B , DU H D , et al . An attention mechanism-guided domain adversarial network for gearbox fault diagnosis under different operating conditions [J]. Transactions of the Institute of Measurement and Control , 2024 , 46 ( 5 ): 927 - 937 .
LI X H , WAN S K , LIU S J , et al . Bearing fault diagnosis method based on attention mechanism and multilayer fusion network [J]. ISA Transactions , 2022 , 128 ( Part B ): 550 - 564 .
DING Y F , JIA M P , ZHUANG J C , et al . Deep imbalanced domain adaptation for transfer learning fault diagnosis of bearings under multiple working conditions [J]. Reliability Engineering & System Safety , 2023 , 230 : 108890 .
PANG S , YANG X Y , ZHANG X F , et al . Fault diagnosis of rotating machinery with ensemble kernel extreme learning machine based on fused multi-domain features [J]. ISA Transactions , 2020 , 98 : 320 - 337 .
TANG Z H , WANG M J , OUYANG T H , et al . A wind turbine bearing fault diagnosis method based on fused depth features in time-frequency domain [J]. Energy Reports , 2022 , 8 : 12727 - 12739 .
LI Y Z , DAI W , ZHANG W F . Bearing fault feature selection method based on weighted multidimensional feature fusion [J]. IEEE Access , 2020 , 8 : 19008 - 19025 .
SENANAYAKA J S L , VAN KHANG H , ROBBERSMYR K G . Multiple classifiers and data fusion for robust diagnosis of gearbox mixed faults [J]. IEEE Transactions on Industrial Informatics , 2019 , 15 ( 8 ): 4569 - 4579 .
LV H X , CHEN J L , PAN T Y , et al . Attention mechanism in intelligent fault diagnosis of machinery:a review of technique and application [J]. Measurement , 2022 , 199 : 111594 .
ZADEH S G , SCHMID M . Bias in cross-entropy-based training of deep survival networks [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence , 2021 , 43 ( 9 ): 3126 - 3137 .
0
浏览量
0
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