吉林大学学报(工学版) ›› 2022, Vol. 52 ›› Issue (12): 2827-2838.doi: 10.13229/j.cnki.jdxbgxb20210415

• 车辆工程·机械工程 • 上一篇    下一篇

基于优化堆叠降噪自编码器的滚动轴承故障诊断

杜先君1,2,3(),贾亮亮1   

  1. 1.兰州理工大学 电气工程与信息工程学院,兰州 730050
    2.兰州理工大学 甘肃省工业过程先进控制重点实验室,兰州 730050
    3.兰州理工大学 国家级电气与控制工程实验教学中心,兰州 730050
  • 收稿日期:2021-05-10 出版日期:2022-12-01 发布日期:2022-12-08
  • 作者简介:杜先君(1979-),男,副教授,博士. 研究方向:复杂系统建模与控制. E-mail:xdu@lut.edu.cn
  • 基金资助:
    国家自然科学基金项目(61963025);甘肃省高等学校创新基金项目(2021A-027)

Fault diagnosis of rolling bearing based on optimized stacked denoising auto encoders

Xian-jun DU1,2,3(),Liang-liang JIA1   

  1. 1.College of Electrical and Information Engineering,Lanzhou University of Technology,Lanzhou 730050,China
    2.Key Laboratory of Gansu Advanced Control for Industrial Processes,Lanzhou University of Technology,Lanzhou 730050,China
    3.National Demonstration Center for Experimental Electrical and Control Engineering Education,Lanzhou University of Technology,Lanzhou 730050,China
  • Received:2021-05-10 Online:2022-12-01 Published:2022-12-08

摘要:

针对深度神经网络用于滚动轴承故障诊断时,网络隐含层层数、各隐含层节点数、稀疏系数以及输入数据置零比例等超参数会直接影响网络诊断性能的问题,提出了一种优化改进堆叠降噪自编码器(SDAE)的滚动轴承故障智能诊断方法。使用学生心理优化算法(SPBO)对降噪自编码器(DAE)网络的超参数进行自适应选取来确定SDAE网络的最优结构和参数,据此提取具有更强表征力的故障状态特征表示,输入到soft-max分类器实现滚动轴承运行工况的精确诊断。使用3个开源数据集对所提网络的性能进行验证,实验结果表明,基于SPBO-SDAE网络的诊断方法在特征有效提取、诊断速度以及故障诊断准确率方面均优于支持向量机(SVM)、反向传播(BP)神经网络、径向基(RBF)神经网络、SDAE网络、SPBO优化后的深度置信网络(DBN)、遗传算法(GA)优化后的SDAE网络以及粒子群算法(PSO)优化后的SDAE网络。

关键词: 故障诊断, 滚动轴承, 堆叠降噪自编码器, 超参优化, 特征提取

Abstract:

In view of the problem that hyper parameters such as the number of hidden layers, the number of nodes in each hidden layer, the sparse coefficient and the dropout ratio of input data will directly affect the diagnosis performance of the network when deep neural network is used for rolling bearing fault diagnosis, an intelligent fault diagnosis method of rolling bearing based on optimized and improved stacked denoising auto encoders (SDAE) was proposed. The student psychology based optimization (SPBO) algorithm was used to adaptively select the hyper parameters of the denoising auto encoder (DAE) network to determine the optimal structure and parameters of SDAE network. Then the fault state features with stronger representation was extracted and input to soft-max classifier to achieve accurate diagnosis of rolling bearing operating conditions. Three open source datasets are used to verify the performance of the proposed network, the results illustrate that the diagnosis method based on SPBO-SDAE network is superior to support vector machine (SVM), back propagation (BP) neural network, radial basis function (RBF) neural network, traditional SDAE network, SPBO based deep belief network (DBN), genetic algorithm (GA) based SDAE network and particle swarm optimization (PSO) based SDAE network in feature extraction, diagnosis speed and fault diagnosis accuracy.

Key words: fault diagnosis, rolling bearing, stacked denoising auto encoders, hyperparametric optimization, feature extraction

中图分类号: 

  • TP277

图1

自编码器网络结构"

图2

DAE网络训练示意图"

图3

不同深度SDAE网络诊断结果"

图4

20次仿真实验结果"

图5

不同种群数量对应的适应度曲线"

图6

SPBO-SDAE网络训练示意图"

图7

SPBO-SDAE网络滚动轴承故障诊断流程图"

表1

CWRU数据集中滚动轴承故障类型"

标签故障类型故障位置故障直径/cm故障程度
1Norm0正常
2B07滚动体0.017 78轻度
3B14滚动体0.035 56中度
4B21滚动体0.053 34重度
5IR07内圈0.017 78轻度
6IR14内圈0.035 56中度
7IR21内圈0.053 34重度
8OR07外圈0.017 78轻度
9OR14外圈0.035 56中度
10OR21外圈0.053 34重度

表2

SPBO优化SDAE网络超参数(CWRU数据集)"

SPBO编码参数符号取值
Xi1第1个隐含层节点数m146
Xi2第2个隐含层节点数m298
Xi3第3个隐含层节点数m332
Xi4第1个隐含层稀疏系数ρ10.2497
Xi5第2个隐含层稀疏系数ρ20.3216
Xi6第3个隐含层稀疏系数ρ30.1415
Xi7输入数据置零比例P0.05

图8

CWRU数据集轴承故障分类混淆矩阵"

表3

PU数据集中滚动轴承故障类型"

标签故障类型故障位置故障描述
1K001正常
2K002正常
3K003正常
4K004正常
5KA01外圈人为损坏
6KA03外圈人为损坏
7KI01内圈人为损坏
8KI03内圈人为损坏
9KA04外圈真实损坏
10KA15外圈真实损坏
11KI04内圈真实损坏
12KI14内圈真实损坏

表4

SPBO优化SDAE网络超参数(PU数据集)"

SPBO编码参数符号取值
Xi1第1个隐含层节点数m183
Xi2第2个隐含层节点数m267
Xi3第3个隐含层节点数m342
Xi4第1个隐含层稀疏系数ρ10.2684
Xi5第2个隐含层稀疏系数ρ20.3317
Xi6第3个隐含层稀疏系数ρ30.1514
Xi7输入数据置零比例P0.04

图9

PU数据集中轴承故障分类混淆矩阵"

表5

MFPT数据集中滚动轴承故障类型"

标签故障类型故障位置负载/kg
1baseline_1122
2baseline_1122
3baseline_1122
4OR_1外圈11
5OR_2外圈23
6OR_3外圈45
7IR_1内圈0
8IR_2内圈23
9IR_3内圈45

表6

SPBO优化SDAE网络超参数(MFPT数据集)"

SPBO编码参数符号取值
Xi1第1个隐含层节点数m161
Xi2第2个隐含层节点数m257
Xi3第3个隐含层节点数m331
Xi4第1个隐含层稀疏系数ρ10.2315
Xi5第2个隐含层稀疏系数ρ20.2706
Xi6第3个隐含层稀疏系数ρ30.2820
Xi7输入数据置零比例P0.07

图10

MFPT数据集中轴承故障分类混淆矩阵"

表7

不同网络故障诊断准确率对比"

网络训练准确率/%测试准确率/%训练时间/s测试时间/s
BP91.1291.036.865.62
RBF92.3191.867.866.43
SVM85.6784.8910.474.35
SPBO-DBN93.2393.1445.378.55
3层SDAE95.9795.1423.985.77
GA-SDAE97.5497.0831.228.96
PSO-SDAE96.8996.2128.357.61
SPBO-SDAE99.8299.3024.315.93

图11

CWRU数据集中测试集误差曲线"

图12

PU数据集中数据测试集误差曲线"

图13

MFPT数据集中测试集误差曲线"

1 Shifat T A, Hur J W. ANN assisted multi sensor information fusion for BLDC motor fault diagnosis[J]. IEEE Access, 2021, 9: 9429-9441.
2 Li J, Yao X, Wang X, et al. Multiscale local features learning based on BP neural network for rolling bearing intelligent fault diagnosis[J]. Measurement, 2019, 153: No.107419.
3 Zhou J, Wang F, Zhang C, et al. Evaluation of rolling bearing performance degradation using wavelet packet energy entropy and RBF neural network[J]. Symmetry, 2019, 11(8): 1064-1074.
4 张鑫, 郭顺生, 李益兵, 等. 基于拉普拉斯特征映射和深度置信网络的半监督故障识别[J]. 机械工程学报, 2020, 56(1): 83-95.
Zhang Xin, Guo Shun-sheng, Li Yi-bing, et al. Semi-supervised fault identification based on laplacian feature map and deep confidence network[J]. Journal of Mechanical Engineering, 2020, 56(1): 83-95.
5 文成林, 吕菲亚. 基于深度学习的故障诊断方法综述[J]. 电子与信息学报, 2020, 42(1): 234-248.
Wen Cheng-lin, Fei-ya Lü. Review of fault diagnosis methods based on deep learning[J]. Journal of Electronics & Information Technology, 2020, 42(1): 234-248.
6 Han S, Oh S, Jeong J. Bearing fault diagnosis based on multiscale convolutional neural network using data augmentation[J]. Journal of Sensors, 2021, 2021(1): 1-14.
7 Feng J, Yao Y, Lu S, et al. Domain knowledge-based deep-broad learning framework for fault diagnosis[J]. IEEE Transactions on Industrial Electronics, 2020, 68(4): 3454-3464.
8 Xing S, Lei Y, Wang S, et al. Distribution-invariant deep belief network for intelligent fault diagnosis of machines under new working conditions[J]. IEEE Transactions on Industrial Electronics, 2020, 68(3): 2617-2625.
9 He J, Ouyang M, Yong C, et al. A novel intelligent fault diagnosis method for rolling bearing based on integrated weight strategy features learning[J]. Sensors, 2020, 20(6): No.1774.
10 Liu S, Jiang H, Wu Z, et al. Rolling bearing fault diagnosis using variational autoencoding generative adversarial networks with deep regret analysis[J]. Measurement, 2021, 168: No.108371.
11 Xu F, Tse P W. Automatic roller bearings fault diagnosis using DSAE in deep learning and CFS algorithm[J]. Soft Computing, 2019, 23(13): 5117-5128.
12 Shao Hai-dong, Jiang Hong-kai, Li Xing-qiu, et al. Intelligent fault diagnosis of rolling bearing using deep wavelet auto-encoder with extreme learning machine[J]. Knowledge-Based Systems, 2018, 140(15): 1-14.
13 Shin H C, Orton M R, Collins D J, et al. Stacked autoencoders for unsupervised feature learning and multiple organ detection in a pilot study using 4D patient data[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2012, 35(8): 1930-1943.
14 雷亚国, 贾峰, 周昕, 等. 基于深度学习理论的机械装备大数据健康监测方法[J]. 机械工程学报, 2015, 51(21): 49-56.
Lei Ya-guo, Jia Feng, Zhou Xin, et al. Big data health monitoring method of mechanical equipment based on deep learning theory[J]. Journal of Mechanical Engineering, 2015, 51(21): 49-56.
15 Suk H I, Lee S W, Shen D. Latent feature representation with stacked auto-encoder for AD/MCI diagnosis[J]. Brain Structure and Function, 2015, 220(2): 841-859.
16 Hong C, Yu J, Wan J, et al. Multimodal deep autoencoder for human pose recovery[J]. IEEE Transactions on Image Processing, 2015, 24(12): 5659-5670.
17 Vincent P, Larochelle H, Lajoie I, et al. Stacked denoising autoencoders:learning useful representations in a deep network with a local denoising criterion[J]. Journal of Machine Learning Research, 2010, 11(12): 3371-3408.
18 Arablouei R, Werner S, Doğançay K. Analysis of the gradient-descent total least-squares adaptive filtering algorithm[J]. IEEE Transactions on Signal Processing, 2014, 62(5): 1256-1264.
19 Das B, Mukherjee V, Das D. Student psychology based optimization algorithm: a new population based optimization algorithm for solving optimization problems[J]. Advances in Engineering Software, 2020, 146: No.102804.
20 Ranaee V, Ebrahimzadeh A, Ghaderi R. Application of the PSO–SVM model for recognition of control chart patterns[J]. ISA Transactions, 2010, 49(4): 577-586.
21 Al-Faisal H R, Ahmad I, Salman A A, et al. Adaptation of population size in sine cosine algorithm[J]. IEEE Access, 2021, 9: 25258-25277.
22 Smith W A, Randall R B. Rolling element bearing diagnostics using the case western reserve university data: a benchmark study[J]. Mechanical Systems and Signal Processing, 2015, 64: 100-131.
23 Hasan B M. Current based condition monitoring of electromechanical systems. Model-free drive system current monitoring: faults detection and diagnosis through statistical features extraction and support vector machines classification[D]. Bradford: School of Engineering Design and Technology, University of Bradford, 2013.
24 Lee D, Siu V, Cruz R, et al. Convolutional neural net and bearing fault analysis[C]∥Proceedings of the International Conference on Data Science (ICDATA). The Steering Committee of The World Congress in Computer Science, Computer Engineering and Applied Computing, Lavasa City, Pune, 2016: 194-200.
25 Zhang X, Li J, Cai Z, et al. Over-fitting suppression training strategies for deep learning-based atrial fibrillation detection[J]. Medical & Biological Engineering & Computing, 2021, 59(1): 165-173.
[1] 欧阳丹彤,孙睿,田新亮,高博涵. 基于集合阻塞的不确定系统中传感器选择方法[J]. 吉林大学学报(工学版), 2023, 53(2): 547-554.
[2] 周怡娜,董宏丽,张勇,路敬祎. 基于VMD去噪和散布熵的管道信号特征提取方法[J]. 吉林大学学报(工学版), 2022, 52(4): 959-969.
[3] 罗巍,卢博,陈菲,马腾. 基于PSO-SVM及时序环节的数控刀架故障诊断方法[J]. 吉林大学学报(工学版), 2022, 52(2): 392-399.
[4] 高文志,王彦军,王欣伟,张攀,李勇,董阳. 基于卷积神经网络的柴油机失火故障实时诊断[J]. 吉林大学学报(工学版), 2022, 52(2): 417-424.
[5] 王进花,胡佳伟,曹洁,黄涛. 基于自适应变分模态分解和集成极限学习机的滚动轴承多故障诊断[J]. 吉林大学学报(工学版), 2022, 52(2): 318-328.
[6] 董绍江,朱朋,裴雪武,李洋,胡小林. 基于子领域自适应的变工况下滚动轴承故障诊断[J]. 吉林大学学报(工学版), 2022, 52(2): 288-295.
[7] 李国发,王彦博,何佳龙,王继利. 机电装备健康状态评估研究进展及发展趋势[J]. 吉林大学学报(工学版), 2022, 52(2): 267-279.
[8] 曹洁,马佳林,黄黛麟,余萍. 一种基于多通道马尔可夫变迁场的故障诊断方法[J]. 吉林大学学报(工学版), 2022, 52(2): 491-496.
[9] 邓飞跃,吕浩洋,顾晓辉,郝如江. 基于轻量化神经网络Shuffle⁃SENet的高速动车组轴箱轴承故障诊断方法[J]. 吉林大学学报(工学版), 2022, 52(2): 474-482.
[10] 张龙,徐天鹏,王朝兵,易剑昱,甄灿壮. 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.
[11] 陈晓雷,孙永峰,李策,林冬梅. 基于卷积神经网络和双向长短期记忆的稳定抗噪声滚动轴承故障诊断[J]. 吉林大学学报(工学版), 2022, 52(2): 296-309.
[12] 曹洁,何智栋,余萍,王进花. 数据不平衡分布下轴承故障诊断方法[J]. 吉林大学学报(工学版), 2022, 52(11): 2523-2531.
[13] 戴礼灿,代翔,崔莹,魏永超. 基于深度集成学习的社交网络异常数据挖掘算法[J]. 吉林大学学报(工学版), 2022, 52(11): 2712-2717.
[14] 陈菲,杨峥,张志成,罗巍. 面向无标签数据的旋转机械故障诊断方法[J]. 吉林大学学报(工学版), 2022, 52(11): 2514-2522.
[15] 欧阳宁,李祖锋,林乐平. 基于多层次空谱融合网络的高光谱图像分类[J]. 吉林大学学报(工学版), 2022, 52(10): 2438-2446.
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
[1] 董婧;赵晓晖;应娜. 基于二进小波变换的基音检测算法[J]. 吉林大学学报(工学版), 2006, 36(06): 978 -0982 .
[2] 何仁, 陈庆樟. 用双开关磁阻电机的汽车能量再生制动技术[J]. 吉林大学学报(工学版), 2009, 39(05): 1137 -1141 .
[3] 李所军,高海波,邓宗全. 为月球车低能耗越障的摇臂悬架参数多目标优化[J]. 吉林大学学报(工学版), 2010, 40(03): 729 -0734 .
[4] 贺宝琴,吴允柱,傅立敏 . 汽车外形对智能车辆队列行驶气动特性的影响[J]. 吉林大学学报(工学版), 2008, 38(01): 7 -011 .
[5] 孙锋, 王殿海, 胡宏宇, 马东方. 基于出行时间分析的公交跳站运行方案优化[J]. 吉林大学学报(工学版), 2012, 42(增刊1): 179 -183 .
[6] 张培林, 陈彦龙, 王怀光, 李胜. 考虑信号特点的合成量子启发结构元素[J]. 吉林大学学报(工学版), 2015, 45(4): 1181 -1188 .
[7] 杨忠振,崔丛 . 基于神经网络的道路交通污染物浓度预测[J]. 吉林大学学报(工学版), 2007, 37(03): 705 -0708 .
[8] 姜辉, 郭孔辉, 张建伟. 基于路径规划的自动平行泊车转向控制器[J]. 吉林大学学报(工学版), 2011, 41(02): 293 -0297 .
[9] 杨庆芳, 张彪, 高鹏. 基于改进动态递归神经网络的交通量短时预测方法[J]. , 2012, 42(04): 887 -891 .
[10] 段跃华, 张肖宁. 基于CT断层扫描图像的混凝土粗集料三维虚拟筛分[J]. , 2012, 42(04): 918 -923 .