改进孪生网络在小样本轴承故障诊断中的应用

doi:10.3778/j.issn.1002-8331.2204-0170

摘要/Abstract

摘要： 针对传统深度学习模型过度依赖大量训练数据，在小样本条件下会出现准确率低和泛化性差的问题，提出一种基于改进孪生神经网络（improved siamese neural network，ISNN）的小样本轴承故障诊断方法。该方法致力于利用度量学习思想来判断样本间的相似性，相比于复杂、多层的深度模型其结构更简单，在小样本下也更容易训练。具体而言，共设计了三个模块来实现故障诊断。特征提取阶段分别利用长短时记忆网络和卷积神经网络提取出故障信号的时间和空间特征，能够充分利用有限的样本信息；关系度量阶段以自适应的网络度量方式来代替常用的欧式距离度量，并且引入全局均值池化来减少网络参数；此外，与标准孪生网络相比，构建的故障分类网络可以直接对样本进行故障识别，在测试阶段无须繁琐的样本比对。实验结果表明，ISNN方法在有限的训练样本数量下，故障诊断准确率和泛化性能均优于各种对比方法。

关键词: 滚动轴承, 孪生网络, 小样本, 故障诊断

Abstract: Aiming at the problems of low accuracy and poor generalization under the condition of small samples due to the excessive dependence of traditional deep learning models on a large amount of training data, a small sample bearing fault diagnosis method based on improved siamese neural network is proposed. This method is dedicated to using the metric learning to judge the similarity between samples. Compared with complex and multi-layer deep models, its structure is simpler and it is easier to train with small samples. Specifically, this paper designs three modules to realize fault diagnosis. In the feature extraction stage, the long short-term memory network and the convolutional neural network are used to extract the temporal and spatial features of the fault signal, which can make full use of the limited sample information; the relationship measurement stage replaces the common Euclidean distance measurement with an adaptive network measurement method and uses global mean pooling to reduce network parameters; in addition, compared with the standard siamese network, the constructed fault classification network can directly identify the faults of the samples, without tedious sample comparison in the testing stage. The experimental results show that the ISNN method has better fault diagnosis accuracy and generalization performance than various comparative methods under the limited number of training samples.

Key words: rolling bearing, siamese neural network, small sample, fault diagnosis

赵晓平, 彭澎, 张永宏, 张中洋. 改进孪生网络在小样本轴承故障诊断中的应用[J]. 计算机工程与应用, 2023, 59(19): 294-304.

ZHAO Xiaoping, PENG Peng, ZHANG Yonghong, ZHANG Zhongyang. Application of Improved Siamese Neural Network in Small Sample Fault Diagnosis of Bearing[J]. Computer Engineering and Applications, 2023, 59(19): 294-304.

参考文献

[1] 莫代一，崔玲丽，王婧.基于双重Q因子的稀疏分解放在滚动轴承早期故障诊断中的引用[J].机械工程学报，2013，24（1）：35-41.
MO D Y，CUI L L，WANG J.Sparse signal decomposition method based on the daul Q-factor and its application to rolling bearing early fault diagnosis[J].Journal of Mechanical Engineering，2013，24（1）：35-41.
[2] LEI Y，HAN D，LIN J，et al.Planetary gearbox fault diagnosis using an adaptive stochastic resonance method[J].Mechanical Systems and Signal Processing，2013，38（1）：113-124.
[3] 谯自健，束学道.非对称势诱导随机共振增强机械重复瞬态提取[J].机械工程学报，2021，57（23）：160-168.
QIAO Z J，SHU X D.Stochastic resonance induced by asymmetric potentials enhanced mechanical repetitive transient extraction[J].Journal of Mechanical Engineering，2021，57（23）：160-168.
[4] 王新，闫文源.基于变分模态分解和SVM的滚动轴承故障诊断[J].振动与冲击，2017，36（18）：252-256.
WANG X，YAN W Y.Fault diagnosis of roller bearings based on the variational model decomposition and SVM[J].Journal of Vibration and Shock，2017，36（18）：252-256.
[5] 蒋永华，程光明，阚君武，等.基于NGA优化SVM的滚动轴承故障诊断[J].仪器仪表学报，2013，34（12）：2684-2689.
JIANG Y H，CHENG G M，KAN J W，et al.Rolling bearing fault diagnosis based on NGA optimized SVM[J].Chinese Journal of Scientific Instrument，2013，34（12）：2684-2689.
[6] WANG F T，LIU X F，DENG G，et al.Remaining life prediction method for rolling bearing based on the long short-term memory network[J].Neural Processing Letters，2019，10：1-18.
[7] 陈保家，陈学力，沈保明，等.CNN-LSTM深度神经网络在滚动轴承故障诊断中的应用[J].西安交通大学学报，2021，55（6）：28-36.
CHEN B J，CHEN X L，SHEN B M，et al.An application of convolution neural network and long short-term memory in rolling bearing fault diagnosis[J].Journal of Xi’an Jiaotong University，2021，55（6）：28-36.
[8] 李恒，张氢，秦仙蓉，等.基于短时傅里叶变换和卷积神经网络的轴承故障诊断方法[J].振动与冲击，2018，37（19）：124-131.
LI H，ZHANG Q，QIN X R，et al.Fault diagnosis method for rolling bearings based on short-time Fourier transform and convolution neural network[J].Journal of Vibration and Shock，2018，37（19）：124-131.
[9] 王丽华，谢阳阳，周子贤，等.基于卷积神经网络的异步电机故障诊断[J].振动、测试与诊断，2017，37（6）：1208-1215.
WANG L H，XIE Y Y，ZHOU Z X，et al.Asynchronous motor fault diagnosis based on convolutional neural network[J].Journal of Vibration，Measurement and Diagnosis，2017，37（6）：1208-1215．
[10] CHEN X H，CHENG G，SHAN XL，et al.Research of weak fault feature information extraction of planetary gear based on ensemble empirical mode decomposition and adaptive stochastic resonance[J].Measurement，2015，73（2）：55-67.
[11] LI F F，FERGUS R，PERONA P.One-shot learning of object categories[J].IEEE Transactions on Pattern Analysis and Machine Intelligence，2006，28（4）：594-611.
[12] 李新叶，龙慎鹏，朱婧.基于深度神经网络的少样本学习综述[J].计算机应用研究，2020（8）：2241-2247.
LI X Y，LONG S P，ZHU J.Survey of few-shot learning based on deep neural network[J].Application Research of Computers，2020（8）：2241-2247.
[13] 吕枫，王义，阮胡林，等.深度嵌入关系空间下齿轮箱标记样本扩充及其半监督故障诊断方法[J].仪器仪表学报，2021，42（2）：55-65.
LV F，WANG Y，RUAN H L，et al.Labeled sample augmentation based on deep embedding relation space for semi-supervised fault diagnosis of gearbox[J].Chinese Journal of Scientific Instrument，2021，42（2）：55-65.
[14] HU T H，TANG T，LIN R L，et al.A simple data augmentation algorithm and a self-adaptive convolutional architecture for few-shot fault diagnosis under different working conditions[J].Measurement，2020，156：107539.
[15] 张西宁，余迪，刘书语.基于迁移学习的小样本轴承故障诊断方法研究[J].西安交通大学学报，2021，55（10）：30-37.
ZHANG X N，YU D，LIU S Y.Research on fault diagnosis method of small sample bearing based on transfer learning[J].Journal of Xi’an Jiaotong University，2021，55（10）：30-37.
[16] 张根保，李浩，冉琰，等.一种用于轴承故障诊断的迁移学习模型[J].吉林大学学报（工学版），2020，50（5）：1617-1626.
ZHANG G B，LI H，RAN Y，et al.A transfer learning model for bearing fault diagnosis[J].Journal of Jilin University（Engineering and Technology Edition），2020，50（5）：1617-1626.
[17] LI X D，HU Y，LI M T，et al.Fault diagnostics between different type of components：a transfer learning approach[J].Applied Soft Computing，2019，86：105950.
[18] 朱瑞金，郝东光，胡石峰.小样本条件下基于卷积孪生网络的变压器故障诊断[J].电力系统及其自动化学报，2021，33（1）：64-68.
ZHU R J，HAO D G，HU S F.Transformer fault diagnosis based on convolutional siamese network with small samples[J].Proceedings of the CSU-EPSA，2021，33（1）：64-68.
[19] 余浩帅，汤宝平，张楷，等.小样本下混合自注意力原型网络的风电齿轮箱故障诊断方法[J].中国机械工程，2021，32（20）：2475-2481.
YU H S，TANG B P，ZHANG K，et al.Fault diagnosis method of wind gearbox mixed with attention prototype network under small sample[J].China Mechanical Engineering，2021，32（20）：2475-2481.
[20] KOCH G，ZEMEL R，SALAKHUTDINOV R.Siamese neural networks for one-shot image recognition[C]//Proceedings of the ICML Deep Learning Workshop，2015.
[21] 樊琳，张惊雷.联合损失优化孪生网络的行人重识别[J].计算机工程与科学，2020，42（2）：273-280．
FAN L，ZHANG J L.Person re-identification based on joint loss and siamese network[J].Computer Engineering & Science，2020，42（2）：273-280.
[22] ZHANG Y，PARDO B，DUAN Z.Siamese style convolutional neural networks for sound search by vocal imitation[J].IEEE/ACM Transactions on Audio，Speech，and Language Processing，2018，27（2）：429-441.
[23] AHRABIAN K，BABAALI B.Usage of autoencoders and siamese networks for online handwritten signature verification[J].Neural Computing and Applications，2019，31（12）：9321-9334.
[24] 赵凯琳，靳小龙，王元卓.小样本学习研究综述[J].软件学报，2021，32（2）：349-369.
ZHAO K L，JIN X L，WANG Y Z.Survey on few-shot learning[J].Journal of Software，2021，32（2）：349-369.
[25] 宋闯，赵佳佳，王康，等.面向智能感知的小样本学习研究综述[J].航空学报，2020，41（S1）：15-28.
SONG C，ZHAO J J，WANG K，et al.A survey of few shot learning based on intelligent perception[J].Acta Aeronautica et Astronautica Sinica，2020，41（S1）：15-28.
[26] CHICCO D.Siamese neural networks：an overview[J].Artificial Neural Networks，2021，2190：73-94.
[27] PASCANU R，MIKOLOV T，BENGIO Y.On thedifficulty of training recurrent neural networks[C]//Proceedings of the 30th International Conference on Machine Learning.Pittsburgh，PA，USA：IMLS，2013：2347-2355.
[28] LIN M，CHEN Q，YAN S.Network in network[J].arXiv：1312.4400，2013.
[29] SNELL J，SWERSKY K，ZEMELN R S.Prototypical networks for few-shot learning[J].arXiv：1703.05175，2017.
[30] 彭成，李凤娟，蒋金元.基于深度学习的故障诊断与预测方法综述[J].现代电子技术，2022，45（3）：111-120.
PENG C，LI F J，JIANG J Y.Summary of fault diagnosis and prediction methods based on deep learning[J].Modern Electronics Technique，2022，45（3）：111-120.