2025年 03期

摘要(Abstract)：

针对工业场景下齿轮箱故障频发、深度学习诊断过程中数据需求量大和可解释性低的问题，提出一种基于决策融合方法和迁移学习的齿轮箱故障诊断模型。基于知识驱动提取CWRU电机轴承故障数据集的振动信号特征，通过递归特征消除算法筛选出最优特征子集，包括复杂包络谱、时域统计特征和小波包分析特征；使用6种分类器并结合投票法、堆叠法和融合法的决策融合方法，建立故障诊断模型；通过基于网络模型的迁移学习，将在CWRU电机轴承故障数据集训练好的模型应用于东南大学齿轮箱轴承数据集，实现对齿轮箱的故障诊断。结果表明，基于堆叠算法构建的6种分类器集成模型在CWRU电机轴承故障数据集和齿轮箱齿轮及轴承故障诊断任务上均表现出优异的性能，且2个任务的诊断准确率差异较小。该模型在齿轮箱齿轮及轴承故障诊断任务的准确率达到100%,与其他故障诊断模型相比，具有较好的故障诊断能力。

关键词(KeyWords)：齿轮箱;故障诊断;特征提取;决策融合;迁移学习

基金项目(Foundation):国家自然科学基金项目(52171185,52371194)

作者(Author): 刘婷婷,王哲铭,于文英,卢武,蔚伟,刘永生

DOI: 10.13349/j.cnki.jdxbn.20250326.002

参考文献(References)：

[1] LIU W Y.Intelligent fault diagnosis of wind turbines using multi-dimensional kernel domain spectrum technique[J].Measurement,2019,133:303.

[2] GAO Q W,LIU W Y,TANG B P,et al.A novel wind turbine fault diagnosis method based on intergral extension load mean decomposition multiscale entropy and least squares support vector machine[J].Renewable Energy,2018,116:169.

[3] XIA H,DAI L,SUN L P,et al.Analysis of the spatiotemporal distribution pattern and driving factors of renewable energy power generation in China[J].Economic Analysis and Policy,2023,80:414.

[4] CHEN R X,HUANG X,YANG L X,et al.Intelligent fault diagnosis method of planetary gearboxes based on convolution neural network and discrete wavelet transform[J].Computers in Industry,2019,106:48.

[5] REN H,LIU W Y,SHAN M C,et al.A novel wind turbine health condition monitoring method based on composite variational mode entropy and weighted distribution adaptation[J].Renewable Energy,2021,168:972.

[6] 王皓，周峰.基于小波包和BP神经网络的风机齿轮箱故障诊断[J].噪声与振动控制，2015,35(2):154.

[7] 王超，李大忠.基于 LSTM 网络的风机齿轮箱轴承故障预警[J].电力科学与工程，2020,36(9):40.

[8] 孟繁晔，高翼飞，陈长征，等.基于多方向振动数据的风机齿轮箱故障智能诊断[J].机械工程师，2022,16:66.

[9] 侯召国，王华伟，王峻洲，等.基于迁移学习与加权多通道融合的齿轮箱故障诊断[J].振动与冲击，2023,42(9):236.

[10] GUO Y L,WU G X,LIU X L.Small sample MKFCNN-LSTM transfer learning fault diagnosis method[C]//ZHANG H,FENG G J,WANG H J,et al.Proceedings of IncoME-VI and TEPEN 2021:Performance Engineering and Maintenance Engineering.Cham:Springer International Publishing,2022:265.

[11] WAN Z T,YANG R,HUANG M J.Deep transfer learning-based fault diagnosis for gearbox under complex working conditions[J].Shock and Vibration,2020,2020:1.

[12] LU Y,TANG J.On time-frequency domain feature extraction of wave signals for structural health monitoring[J].Measurement,2018,114:51.

[13] XIA Z G,XIA S X,WAN L,et al.Spectral regression based fault feature extraction for bearing accelerometer sensor signals[J].Sensors,2012,12(10):13694.

[14] TOMA R N,PROSVIRIN A E,KIM J M.Bearing fault diagnosis of induction motors using a genetic algorithm and machine learning classifiers[J].Sensors,2020,20(7):1884.

[15] HU D,ZHANG C,YANG T,et al.An intelligent anomaly detection method for rotating machinery based on vibration vectors[J].IEEE Sensors Journal,2022,22(14):14294.

[16] ZHANG Y F,KANG B Y,HOOI B,et al.Deep long-tailed learning:a survey[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2023,45(9):10795.

[17] RAFIEE R,RAFIEE S,TSE P W.Bearing fault diagnosis based on wavelet transform and fuzzy logic[J].Mechanical Systems and Signal Processing,2011,25(2):667.

[18] ROSENSTEIN M T,MARX Z,KAELBLING L P,et al.To transfer or not to transfer[C]//NIPS 2005 Workshop on Inductive Transfer:10 Years Later,December 5-8,2005,Vancouver,Canada.Cambridge:MIT Press 2005,898(3):4.

[19] AGAHI H,MAHMOODZADEH A.Decision fusion scheme for bearing defects diagnosis in induction motors[J].Electrical Engineering,2020,102(4):2269.

[20] AJAKAN H,GERMAIN P,LAROCHELLE H,et al.Domain-adversarial neural networks[EB/OL].(2014-12-15) [2024-02-01].https://doi.org/10.48550/arXiv.1412.4446.

[21] HINTON G,VINYALS O,DEAN J.Distilling the knowledge in a neural network[EB/OL].(2015-03-09)[2024-02-01].https://doi.org/10.48550/arXiv.1503.02531.

[22] SHAO S Y,McALEER S,YAN R Q,et al.Highly accurate machine fault diagnosis using deep transfer learning[J].IEEE Transactions on Industrial Informatics,2019,15(4):2446.

[23] DAI W Y,YANG Q,XUE G R,et al.Boosting for transfer learning[C]//GHAHRAMANI Z.ICML’07:Proceedings of the 24th International Conference on Machine Learning.New York:Association for Computing Machinery,2007:193.

[24] TZENG E,HOFFMAN J,ZHANG N,et al.Deep domain confusion:maximizing for domain invariance[EB/OL].(2014-12-10) [2024-02-01].https://doi.org/10.48550/arXiv.1412.3474.

[25] HUANG J T,LI J Y,YU D,et al.Cross-language knowledge transfer using multilingual deep neural network with shared hidden layers[C]//2013 IEEE International Conference on Acoustics,Speech and Signal Processing,May 26-31,2013,Vancouver,Canada.New York:IEEE,2013:7304.