Improved LeNet-5 Convolutional Neural Network and Application on Mechanical Fault Diagnosis
-
摘要: 针对往复机械设备故障诊断容易受到转速波动影响和深度网络诊断模型复杂、鲁棒性差的问题, 提出了一种一维改进型LeNet-5的机械故障诊断方法, 并对比分析了滑动窗和阶次采样的数据样本构造方法的效果。在经典模型LeNet-5基础上构建了结构简单紧凑的一维卷积神经网络诊断模型, 模型仅包含了两个卷积模块、单一全连接层和输出层, 卷积模块结合批规范化层和Relu层, 提高训练速度和网络泛化能力, 利用重叠池化窗口和随机失活来缓解网络出现过拟合现象。利用凯斯西储大学开源轴承数据集进行验证, 12种工况下的故障识别率能够达到99.82%。针对往复机械的转速波动性影响, 采用阶次采样的数据样本构建方法, 提高网络模型的训练样本数据质量, 柴油机阶次采样条件下可以实现小样本条件下取得良好的训练效果。Abstract: Aiming at the problems that the fault diagnosis of reciprocating machinery equipment is easily affected by speed fluctuations and the deep network diagnosis model is complicated and poor in robustness, a one-dimensional improved LeNet-5 mechanical fault diagnosis method was proposed, and the effects of sliding window and order sampling data sample construction method are compared and analyzed. Based on the classic model LeNet-5, a simple and compact one-dimensional convolutional neural network diagnosis model was constructed, which only contains two convolution modules, a single fully connected layer and an output layer. And its convolution module combines the batch normalization layer and Relu layers to improve training speed and network generalization ability, and use overlapping pooling windows and random inactivation to alleviate network overfitting. Using the open source bearing data set of Case Western Reserve University to verify, the fault detection accuracy under 12 working conditions can reach 99.82%. Aiming at the influence of speed fluctuation of reciprocating machinery, the method of constructing data samples of order sampling is adopted to improve the quality of training sample data of the network model. Under the condition of order sampling of diesel engine, good training results can be achieved under the condition of small samples.
-
Key words:
- fault diagnosis /
- convolutional neural network /
- order sampling /
- reciprocating machinery
-
表 1 LeNet-5经典网络结构及参数
网络层 特殊设定 训练参数数量 输出格式 输入层 32×32像素黑白图像 0 32×32×1 卷积层1 6个卷积核, 5×5, 步长1 156 28×28×6 池化层1 池化2×2, 步长2 12 14×14×6 卷积层2 16个卷积核, 5×5, 步长1 1 516 10×10×16 池化层2 池化2×2, 步长2 32 5×5×16 全连接层1 120个神经元 48 120 1×1×120 全连接层2 84个神经元 10 164 1×84 输出层 10个神经元 840 1×10 表 2 改进一维LeNet-5网络结构
网络层 特殊设定 Padding 训练参数数量 输出格式 输入层 一维信号2048 - 0 2 048×1 卷积层1 16个卷积核, 64×1, 步长10 “same” 1 040 256×16 池化层1 池化3×1, 步长2 [0 0 0 0] 32 102×16 卷积层2 32个卷积核, 16×1, 步长4 “same” 8 224 26×32 池化层2 池化3×1, 步长2 [0 0 0 0] 64 12×32 全连接层1 12个神经元 - 4 620 1×12 表 3 数据集与标签
数据 代码 数据 代码 Normal 1 IR014 7 B007 2 IR021 8 B014 3 IR021 9 B021 4 OR007 10 B028 5 OR014 11 IR007 6 OR021 12 -
[1] 王国彪, 何正嘉, 陈雪峰, 等. 机械故障诊断基础研究"何去何从"[J]. 机械工程学报, 2013, 49(1): 63-72 https://www.cnki.com.cn/Article/CJFDTOTAL-JXXB201301010.htmWANG G B, HE Z J, CHEN X F, et al. Basic research on machinery fault diagnosis-what is the prescription[J]. Journal of Mechanical Engineering, 2013, 49(1): 63-72 (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-JXXB201301010.htm [2] 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 https://www.cnki.com.cn/Article/CJFDTOTAL-JXXB201521007.htmLEI Y G, JIA F, ZHOU X, et al. A deep learning-based method for machinery health monitoring with big data[J]. Journal of Mechanical Engineering, 2015, 51(21): 49-56 (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-JXXB201521007.htm [3] HE K M, ZHANG X Y, REN S Q, et al. Spatial pyramid pooling in deep convolutional networks for visual recognition[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2015, 37(9): 1904-1916 doi: 10.1109/TPAMI.2015.2389824 [4] KRIZHEVSKY A, SUTSKEVER I, HINTON G E. ImageNet classification with deep convolutional neural networks[J]. Communications of the ACM, 2017, 60(6): 84-90 doi: 10.1145/3065386 [5] KHAN S, YAIRI T. A review on the application of deep learning in system health management[J]. Mechanical Systems and Signal Processing, 2018, 107: 241-265 doi: 10.1016/j.ymssp.2017.11.024 [6] GUO S, YANG T, GAO W, et al. A novel fault diagnosis method for rotating machinery based on a convolutional neural network[J]. Sensors, 2018, 18(5): 1429 doi: 10.3390/s18051429 [7] 吴定海, 任国全, 王怀光, 等. 基于卷积神经网络的机械故障诊断方法综述[J]. 机械强度, 2020, 42(5): 1024-1032 https://www.cnki.com.cn/Article/CJFDTOTAL-JXQD202005002.htmWU D H, REN G Q, WANG H G, et al. The review of mechanical fault diagnosis methods based on convolutional neural network[J]. Journal of Mechanical Strength, 2020, 42(5): 1024-1032 (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-JXQD202005002.htm [8] WEN L, LI X Y, GAO L, et al. A new convolutional neural network-based data-driven fault diagnosis method[J]. IEEE Transactions on Industrial Electronics, 2018, 65(7): 5990-5998 doi: 10.1109/TIE.2017.2774777 [9] GUO S, YANG T, GAO W, et al. A novel fault diagnosis method for rotating machinery based on a convolutional neural network[J]. Sensors, 2018, 18(5): 1429 doi: 10.3390/s18051429 [10] ZHANG W, LI C H, PENG G L, et al. A deep convolutional neural network with new training methods for bearing fault diagnosis under noisy environment and different working load[J]. Mechanical Systems and Signal Processing, 2018, 100: 439-453 doi: 10.1016/j.ymssp.2017.06.022 [11] WAN L J, CHEN Y W, LI H Y, et al. Rolling-element bearing fault diagnosis using improved LeNet-5 network[J]. Sensors, 2020, 20(6): 1693 doi: 10.3390/s20061693 [12] 刘星辰, 周奇才, 赵炯, 等. 一维卷积神经网络实时抗噪故障诊断算法[J]. 哈尔滨工业大学学报, 2019, 51(7): 89-95 https://www.cnki.com.cn/Article/CJFDTOTAL-HEBX201907013.htmLIU X C, ZHOU Q C, ZHAO J, et al. Real-time and anti-noise fault diagnosis algorithm based on 1-D convolutional neural network[J]. Journal of Harbin Institute of Technology, 2019, 51(7): 89-95 (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-HEBX201907013.htm [13] JANSSENS O, SLAVKOVIKJ V, VERVISCH B, et al. Convolutional neural network based fault detection for rotating machinery[J]. Journal of Sound and Vibration, 2016, 377: 331-345 doi: 10.1016/j.jsv.2016.05.027 [14] LECUN Y, BOTTOU L, BENGIO Y, et al. Gradient- based learning applied to document recognition[J]. Proceedings of the IEEE, 1998, 86(11): 2278-2324 doi: 10.1109/5.726791 [15] LOPAROK A. Bearing data center[EB/OL]. Case Western Reserve University. (2020-09-15) https://engineering.case.edu/bearingdatacenter