一种有效的不均衡样本生成方法及其在行星变速箱故障诊断中的应用

doi:10.3969/j.issn.1000-1093.2019.07.003

兵工学报 ›› 2019, Vol. 40 ›› Issue (7): 1349-1357.doi: 10.3969/j.issn.1000-1093.2019.07.003

一种有效的不均衡样本生成方法及其在行星变速箱故障诊断中的应用

吴春志，冯辅周，吴守军，陈汤，江鹏程

(陆军装甲兵学院车辆工程系，北京 100072)

收稿日期:2018-11-02 修回日期:2018-11-02 上线日期:2019-09-03
通讯作者: 冯辅周(1971—)，男，教授，博士生导师 E-mail:fengfuzhou@tsinghua.org.cn
作者简介:吴春志(1991—)，男，博士研究生。 E-mail: chunzhi.wu@qq.com

An Effective Method for Imbalanced Sample Generation and Its Application in Fault Diagnosis of Planetary Gearbox

WU Chunzhi, FENG Fuzhou, WU Shoujun, CHEN Tang, JIANG Pengcheng

(Department of Mechanical Engineering, Academy of Army Armored Force, Beijing 100072, China)

Received:2018-11-02 Revised:2018-11-02 Online:2019-09-03

摘要/Abstract

摘要： 针对实际运行中行星变速箱故障数据较少、各个状态样本不均衡的问题，提出了由Wasserstein生成式对抗网络(WGAN)样本生成模型和卷积神经网络(CNN)分类模型组合的WGAN-CNN故障诊断分类模型。该模型对故障数据的频谱信号进行过采样，以扩展故障样本数量，从而更好地对故障状态进行分类。采用加州大学欧文分校人工数据集对WGAN生成模型以及经典过采样方法进行对比，并在行星变速箱故障试验台上进行验证。结果表明，样本不均衡会严重影响分类结果，而WGAN-CNN模型可以很好地扩充故障样本集，提高在故障样本稀少情况下的诊断准确率。

关键词: 行星变速箱, 样本不均衡, Wasserstein生成式对抗网络, 卷积神经网络

Abstract: A fault diagnosis classification model based on WGAN-CNN is proposed for few fault data of planetary gearbox and the imbalanced samples of each state in actual operation. The proposed model is a combination of Wasserstein generative adversarial network (WGAN), a sample generation model, a convolutional neural network (CNN), and a sample classification model. The model is used to oversample the spectral signals of fault data and expand the number of fault samples, thus classifying the fault states better. UCI artificial datasets were used to compare WGAN generation model and classical oversampling methods, and verified on a planetary gearbox fault test rig. The results show that the imbalanced samples seriously affect the classification results, and the WGAN-CNN model can well expand the fault sample datasets and improve the diagnostic accuracy in the case of rare fault samples. Key

Key words: planetarygearbox, imbalancesamples, Wassersteingenerativeadversarialnetwork, convolutionalneuralnetwork

中图分类号:

吴春志，冯辅周，吴守军，陈汤，江鹏程. 一种有效的不均衡样本生成方法及其在行星变速箱故障诊断中的应用[J]. 兵工学报, 2019, 40(7): 1349-1357.

WU Chunzhi, FENG Fuzhou, WU Shoujun, CHEN Tang, JIANG Pengcheng. An Effective Method for Imbalanced Sample Generation and Its Application in Fault Diagnosis of Planetary Gearbox[J]. Acta Armamentarii, 2019, 40(7): 1349-1357.

参考文献

［1］ SAWALHI N, RANDALL R B. Gear parameter identification in a wind turbine gearbox using vibration signals[J]. Mechanical Systems & Signal Processing, 2014, 42(1/2):368-376.
[2] 杨大为, 赵永东, 冯辅周, 等.基于参数优化变分模态分解和多尺度熵偏均值的行星变速箱故障特征提取[J].兵工学报,2018,39(9):1683-1691.
YANG D W, ZHAO Y D, FENG F Z, et al. Planetary gearbox fault feature extraction based on parameter optimized variational mode decomposition and partial mean of multi-scale entropy[J]. Acta Armamentarii, 2018,39(9):1683-1691. (in Chinese)
[3] 丁闯, 张兵志, 冯辅周, 等. 局部均值分解和排列熵在行星齿轮箱故障诊断中的应用[J]. 振动与冲击, 2017, 36(17):55-60.
DING C, ZHANG B Z, FENG F Z, et al. Application of local mean decomposition and permutation entropy in fault diagnosis of planetary gearboxes[J]. Journal of Vibration and Shock, 2017, 36(17):55-60. (in Chinese)
[4] DOUZAS G, BACAO F. Effective data generation for imbalanced learning using conditional generative adversarial networks[J]. Expert Systems with Applications, 2018, 91:464-471.
[5] 杨宇, 潘海洋, 李永国, 等. 一种增量式半监督VPMCD齿轮故障在线诊断方法[J]. 振动与冲击, 2015,34(8):49-54.
YANG Y, PAN H Y, LI Y G, et al. A novel incremental semi-supervised VPMCD gear fault on-line diagnosis method[J]. Journal of Vibration and Shock, 2015,34(8):49-54. (in Chinese)
[6] YANG Q, CHEN G M, HE QＦ, et al. The research of pattern recognition of gear pump based on EMD and KPCA-SVM[C]∥Proceedings of International Conference on System Science, Engineering Design and Manufacturing Informatization. Guiyang, China: International Conference on System Science, 2011:1104-1109.
[7] GOLAFSHAN R, SANLITURK K Y. SVD and Hankel matrix based de-noising approach for ball bearing fault detection and its assessment using artificial faults[J]. Mechanical Systems & Signal Processing, 2016, 70:36-50.
[8] 张文斌, 苏艳萍, 普亚松, 等. 基于集合经验模式分解能量分布与灰色相似关联度的齿轮故障诊断[J]. 机械工程学报, 2014, 50(7):70-77.
ZHANG W B, SU Y P, PU Y S, et al. Gear fault diagnosis method using ensemble empirical mode decomposition energy distribution and grey similar incidence[J]. Journal of Mechanical Engineering, 2014, 50(7):70-77. (in Chinese)
[9] GOODFELLOW I J, POUGETABADIE J, MIRZA M, et al. Generative adversarial networks[J]. Advances in Neural Information Processing Systems, 2014, 3:2672-2680.

[10] JING L Y, ZHAO M, LI P, et al. A convolutional neural network based feature learning and fault diagnosis method for the condition monitoring of gearbox[J]. Measurement, 2017, 111:1-10.
[11] INCE T, KIRANYAZ S, EREN L, et al. Real-time motor fault detection by 1-D convolutional neural networks[J]. IEEE Tran- sactions on Industrial Electronics, 2016, 63(11):7067-7075.
[12] CHAWLA N V, BOWYER K W, HALL L O, et al. SMOTE: synthetic minority over-sampling technique[J]. Journal of Artificial Intelligence Research, 2002, 16(1) : 321-357．
[13] BUNKHUMPORNPAT C, SINAPIROMSARAN K, LURSINSAP C. Safe-level-SMOTE: safe-level-synthetic minority over-sampling technique for handling the class imbalanced problem[C]∥Proceedings of Pacific-Asia Conference on Advances in Knowledge Discovery and Data Mining. Hyderabad, India: Springer-Verilog, 2009:475-482.
[14] HAN H, WANG W Y, MAO B H. Borderline-SMOTE: a new over-sampling method in imbalanced data sets learning[C]∥Advances in Intelligent Computing, International Conference on Intelligent Computing. Berlin, Germany: Springer, 2005:878-887.
[15] CIESLAK D A, CHAWLA N V, STRIEGEL A. Combating imbalance in network intrusion datasets[C]∥Proceedings of International Conference on Granular Computing. Atlanta, GA, US: IEEE, 2006:732-737.

第40卷
第7期2019 年7月兵工学报ACTA
ARMAMENTARIIVol.40No.7Jul.2019

[1]	袁航，罗迎，李开明，陈怡君，张群. 基于涡旋电磁波雷达的人体目标步态精细识别[J]. 兵工学报, 2022, 43(5): 1167-1174.
[2]	普运伟，刘涛涛，郭江，吴海潇. 基于卷积神经网络和模糊函数主脊坐标变换的雷达辐射源信号识别[J]. 兵工学报, 2021, 42(8): 1680-1689.
[3]	马月红，孔梦瑶. 基于改进快速区域卷积网络的目标检测轻量化算法[J]. 兵工学报, 2021, 42(12): 2664-2674.
[4]	张宏鹏，黄长强，唐上钦，轩永波. 基于卷积神经网络的无人作战飞机飞行轨迹实时预测[J]. 兵工学报, 2020, 41(9): 1894-1903.
[5]	刘秋，孙晋伟，张华, 胡煦，顾亮. 基于卷积神经网络的路面识别及半主动悬架控制[J]. 兵工学报, 2020, 41(8): 1483-1493.
[6]	呙鹏程，吴礼洋. 融合卷积特征与判别字典学习的低截获概率雷达信号识别[J]. 兵工学报, 2019, 40(9): 1881-1889.
[7]	梁杰，李磊，任君，齐航，周红丽. 基于深度学习的红外图像遮挡干扰检测方法[J]. 兵工学报, 2019, 40(7): 1401-1410.
[8]	张玉燕，李永保，温银堂，张芝威. 基于Faster R-卷积神经网络的金属点阵结构缺陷识别方法[J]. 兵工学报, 2019, 40(11): 2329-2335.
[9]	杨大为，赵永东，冯辅周，江鹏程，丁闯. 基于参数优化变分模态分解和多尺度熵偏均值的行星变速箱故障特征提取[J]. 兵工学报, 2018, 39(9): 1683-1691.
[10]	丁闯，冯辅周，张兵志，吴守军. 行星变速箱振动信号的线性量子信息熵特征[J]. 兵工学报, 2018, 39(12): 2306-2312.
[11]	孙皓泽，常天庆，王全东，孔德鹏，戴文君. 一种基于分层多尺度卷积特征提取的坦克装甲目标图像检测方法[J]. 兵工学报, 2017, 38(9): 1681-1691.
[12]	!:杨绍奎，张福田;2:孟惠荣. 履带车辆行星变速装置复合框架强度可靠度计算的智能仿真算法[J]. 兵工学报, 1999, 20(3): 198-203.

一种有效的不均衡样本生成方法及其在行星变速箱故障诊断中的应用

An Effective Method for Imbalanced Sample Generation and Its Application in Fault Diagnosis of Planetary Gearbox

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 12

编辑推荐

Metrics

本文评价