基于注意力机制改进残差神经网络的轴承故障诊断方法

PDF(3213 KB)

振动与冲击 ›› 2023, Vol. 42 ›› Issue (16) : 82-91.

论文

韩争杰1，牛荣军1，马子魁2，崔永存1，邓四二1

作者信息 +

Bearing fault diagnosis methods based on an attentional-mechanism-improved residual neural network

HAN Zhengjie1，NIU Rongjun1，MA Zikui2，CUI Yongcun1，DENG Sier1

Author information +

文章历史 +

摘要

针对滚动轴承在不同工况环境中故障诊断训练时间长、准确率低和泛化性能弱的问题，提出了基于注意力机制改进残差神经网络的轴承故障诊断方法。为了提高ResNet模型的准确率和泛化性，提出了基于注意力机制的SE-ResNet模型和CBAM-ResNet模型，并在凯斯西储大学(CWRU）数据集上进行了试验，在同工况有训练集的情况下ResNet模型测试的准确率为97.28%，在不同工况下模型直接迁移的准确率为94.14%～96.86%，CBAM-ResNet模型在不同工况下模型直接迁移的准确率为97.14%～98.86%，SE-ResNet模型在不同工况下模型直接迁移的准确率为97.86%～99.71%，两种改进模型的准确率都明显优于原ResNet模型，表明提出的优化模型提高了ResNet模型的准确率和泛化性。

Abstract

Aiming at the problems of long training time, low accuracy and weak generalization performance of rolling bearing fault diagnosis in different working conditions, two bearing fault diagnosis methods based on attention mechanism and improved residual neural network were proposed. In order to improve the accuracy and generalization of the ResNet model, the SE-ResNet model and CBAM-ResNet model based on the attention mechanism were proposed and tested on the Case Western Reserve University (CWRU) data set. The accuracy of the ResNet model was 97.28% under the same working condition with training set. The accuracy of direct model migration under different working conditions is 94.14%-96.86%, that of CBAM-ResNet model under different working conditions is 97.14%-98.86%, that of SE-ResNet model under different working conditions is 97.86%-99.71%. The accuracy of the two improved models is significantly better than that of the original ResNet model, indicating that the proposed optimization model improves the accuracy and generalization of ResNet model.

导出引用

韩争杰1，牛荣军1，马子魁2，崔永存1，邓四二1. 基于注意力机制改进残差神经网络的轴承故障诊断方法[J]. 振动与冲击, 2023, 42(16): 82-91

HAN Zhengjie1，NIU Rongjun1，MA Zikui2，CUI Yongcun1，DENG Sier1. Bearing fault diagnosis methods based on an attentional-mechanism-improved residual neural network[J]. Journal of Vibration and Shock, 2023, 42(16): 82-91

参考文献

[1] 雷亚国,贾峰,孔德同,等. 大数据下机械智能故障诊断的机遇与挑战[J]. 机械工程学报,2018,54(5):94-104.
Lei Yaguo, Jia Feng, Kong Detong, et al. Opportunities and Challenges of Mechanical Intelligent Fault Diagnosis Based on Big Data[J]. Journal of Mechanical Engineering,2018,54
(5):94-104.
[2] 唐斯,陈新楚,郑松. 基于注意力与多尺度卷积神经网络的电机轴承故障诊断[J]. 电气技术,2020,21(11):32-38.
Tang Si, Chen Xin-chu, Zheng Song. Fault Diagnosis of Motor Bearing Based on Attention and Multi-Scale Convolutional Neural Network[J]. Electrical Technology,2020,21(11):32-38.
[3] 雷亚国,何正嘉,訾艳阳,等.基于特征评估和神经网络的机械故障诊断模型[J].西安交通大学学报, 2006, 40(5):558-562.
LEI Ya-guo, HE Zheng-jia, ZI Yan-yang, et al.Mechanical fault diagnosis model based on feature evaluation and neural network[J]. Journal of Xi'an Jiaotong University,2006, 40(5):558-562.
[4] Lei Y , Yang B , Jiang X , et al. Applications of machine learning to machine fault diagnosis: A review and roadmap[J]. Mechanical Systems and Signal Processing, 2020, 138:106587.
[5] 周小龙,刘薇娜,姜振海,等.基于改进HHT和马氏距离的齿轮故障诊断[J].振动与冲击,2017,36(22):218-224.
ZHOU Xiao-long, LIU Wei-na, JIANG Zhen-hai, et al.Gear fault diagnosis based on improved HHT and Mahalanobis distance[J]. Journal of Vibration and Shock,2017,36 (22):218-224.
[6] 陈俊洵,程龙生,胡绍林,等.基于EMD的改进马田系统的滚动轴承故障诊断[J].振动与冲击,2017,36(5):151-156.
CHEN Jun-xun, CHENG Long-sheng, HU Shao-lin, et al. Rolling bearing fault diagnosis based on EMD improved Matte system[J].Journal of Vibration and Shock,2017,36 (5):151-156
[7] Min X, Teng L , Lin X, et al. Fault Diagnosis for Rotating Machinery Using Multiple Sensors and Convolutional Neural Networks[J]. IEEE/ASME Transactions on Mechatronics, 2017, PP(99):1-1.
[8] 姜景升,王华庆,柯燕亮,向伟.基于LTSA与K-最近邻分类器的故障诊断[J].振动与冲击,2017,36(11):134-139.
JIANG Jing-sheng, WANG Hua-qing, KE Yan-liang, Xiang Wei.Fault diagnosis based on LTSA and K-nearest Neighbor classifier[J]. Journal of Vibration and Shock,2017, 36(11):134-139.
[9] 何大伟,彭靖波,胡金海,等.基于改进FOA优化的CS-SVM轴承故障诊断研究[J].振动与冲击,2018,37(18):108-114.
HE Da-wei, PENG Jing-bo, HU Jin-hai, et al.Research on Bearing Fault Diagnosis based on Improved FOA Optimization of CS-SVM[J]. Journal of Vibration and Shock,2018,37(18): 108-114.
[10] 李锋,汤宝平,郭胤. Laplacian双联最小二乘支持向量机用于早期故障诊断[J].振动与冲击,2017,36(16):85-92.
LI Feng, TANG Bao-ping, GUO Yin.Dual least-squares support vector machines are used for early fault diagnosis[J].Journal of Vibration and Shock,2017,36 (16) :85-92.
[11] Ding X, He Q. Energy-Fluctuated Multiscale Feature Learning With Deep ConvNet for Intelligent Spindle Bearing Fault Diagnosis[J]. IEEE Transactions on Instrumentation and Measurement, 2017, (8):1-10.
[12] Yuan M, Wu Y, Lin L. Fault diagnosis and remaining useful life estimation of aero engine using LSTM neural network[C]. 2016 IEEE international conference on aircraft utility systems (AUS). IEEE, 2016: 135-140.
[13] 雷亚国,杨彬,杜兆钧,等.大数据下机械装备故障的深度迁移诊断方法[J].机械工程学报,2019,55(7):1-8.
LEI Ya-guo, YANG Bin, DU Zhao-jun, et al.Deep migration diagnosis method of mechanical equipment fault based on big data[J].Chinese Journal of Mechanical Engineering,2019,55 (7):1-8.
[14] Shao S, McAleer S, Yan R, et al. Highly accurate machine fault diagnosis using deep transfer learning[J]. IEEE Transactions on Industrial Informatics, 2018, 15(4):2446- 2455.
[15] Ma P, Zhang H, Fan W, et al. A novel bearing fault diagnosis method based on 2D image representation and transfer learning-convolutional neural network[J]. Measurement Science and Technology, 2019,30(5): 055402.
[16] Zhu Y, Zhuang F, Wang D. Aligning domain-specific distribution and classifier for cross-domain classification from multiple sources[C]. Proceedings of the AAAI Conference on Artificial Intelligence. 2019, 33(01): 5989-5996.
[17] Li J, Xu Z, Yongkang W, et al. GradMix: Multi-source transfer across domains and tasks[C]. Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision. Snowmass : IEEE ,2020: 3019-3027.
[18] He K, Zhang X, Ren S, et al. Deep residual learning for image recognition[C]. Proceedings of the IEEE conference on computer vision and pattern recognition. Las Vegas: IEEE ,2016: 770-778.
[19] Hu J, Shen L, ALBANIE S, et al. Squeeze-and-excitation networks[C]. Proceedings of the IEEE conference on computer vision and pattern recognition. 2018: 7132-7141.
[20] Woo S, Park J, Lee J Y, et al. Cbam: Convolutional block attention module[C]. Proceedings of the European conference on computer vision (ECCV). 2018: 3-19.