基于多尺度注意力倒残差网络的轴承故障诊断

doi:10.16088/j.issn.1001-6600.2022102601

摘要/Abstract

摘要： 传统应用于滚动轴承故障识别的深度学习模型诊断精度不高,且训练时间长,为了克服上述问题,本文提出多尺度注意力倒残差网络模型(MARCNN)。首先,该模型构造多尺度特征提取模块,采用多尺度并行卷积获取原始信号的不同层次特征,自适应提取故障特征信息;然后,利用特征图膨胀标准卷积构建浅层卷积模块,提高浅层网络学习能力;最后,构造SE-Mobile模块挖掘深层故障特征并缩减模型参数量。该模型融合注意力机制,整合不同维度特征权重,提升模型故障分类能力。采用凯斯西储大学轴承数据集验证该模型性能,实验结果表明,在无噪音、高斯噪声以及变负载工况下准确率分别达到99.98%、98.41%及94.98%,说明该模型具有较好的抗噪能力和泛化性能。

关键词: 注意力机制, 倒残差, 故障诊断, 滚动轴承, 卷积神经网络

Abstract: The traditional deep learning model network applied to rolling bearing fault identification has low diagnosis accuracy and long training time. To overcome these disadvantages, this paper proposes a multi-scale attention reverse residuals convolutional neural network (MARCNN). Firstly, multi-scale feature extraction module adopts multi-scale convolution to obtain different levels of original signal features and adaptively extract fault feature information. Secondly, feature graph expansion standard convolution is used to construct the shallow convolution module to improve the shallow network learning ability. Finally, SE-Mobile module is constructed to mine deep fault features and reduce the number of model parameters. All modules combine attention mechanism to integrate feature weights of different dimensions to improve model feature learning ability. The experimental results show that the model can reduce the number of parameters and improve the training speed, and the accuracy of the model can reach 99.98%, 98.41% and 94.98% under fixed, Gaussian noise and variable load conditions, respectively. It is shown that the model has a certain anti-noise and generalization ability.

Key words: attentional mechanism, reverse residual module, fault diagnosis, rolling bearing, CNN

中图分类号: TP18; TH133.33

唐侯清, 辛斌斌, 朱虹谕, 乙加伟, 张冬冬, 武新章, 双丰. 基于多尺度注意力倒残差网络的轴承故障诊断[J]. 广西师范大学学报（自然科学版）, 2023, 41(4): 109-122.

TANG Houqing, XIN Binbin, ZHU Hongyu, YI Jiawei, ZHANG Dongdong, WU Xinzhang, SHUANG Feng. Rolling Bearing Fault Diagnosis Based on Multi-scale Attentional Inverted Residual Model[J]. Journal of Guangxi Normal University(Natural Science Edition), 2023, 41(4): 109-122.

参考文献

[1] 赵小强, 张亚洲. 利用改进卷积神经网络的滚动轴承变工况故障诊断方法[J]. 西安交通大学学报, 2021, 55(12): 108-118. DOI: 10.7652/xjtuxb202112013.
[2] 赵凯辉, 吴思成, 李涛, 等. 基于Inception-BLSTM的滚动轴承故障诊断方法研究[J]. 振动与冲击, 2021, 40(17):
290-297. DOI: 10.13465/j.cnki.jvs.2021.17.038.
[3] 王贡献, 张淼, 胡志辉, 等. 基于多尺度均值排列熵和参数优化支持向量机的轴承故障诊断[J]. 振动与冲击, 2022, 41(1): 221-228. DOI: 10.13465/j.cnki.jvs.2022.01.028.
[4] 李兵, 韩睿, 何怡刚, 等. 改进随机森林算法在电机轴承故障诊断中的应用[J]. 中国电机工程学报, 2020, 40(4): 1310-1319, 1422. DOI: 10.13334/j.0258-8013.pcsee.190501.
[5] SU Z Q,TANG B P,MA J H,et al. Fault diagnosis method based on incremental enhanced supervised locally linear embedding and adaptive nearest neighbor classifier[J]. Measurement, 2014, 48: 136-148. DOI: 10.1016/j.measurement. 2013.10.041.
[6] DENG F Y, DING H, YANG S P, et al. An improved deep residual network with multiscale feature fusion for rotating machinery fault diagnosis[J]. Measurement Science and Technology, 2020, 32(2): 024002. DOI: 10.1088/1361-6501/abb917.
[7] 袁建虎, 韩涛, 唐建, 等. 基于小波时频图和CNN的滚动轴承智能故障诊断方法[J]. 机械设计与研究, 2017, 33(2): 93-97. DOI: 10.13952/j.cnki.jofmdr.2017.0115.
[8] LIU C, CHENG G, CHEN X H, et al. Planetary gears feature extraction and fault diagnosis method based on VMD and CNN[J]. Sensors, 2018, 18(5): 1523. DOI: 10.3390/s18051523.
[9] 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. DOI: 10.1016/j.measurement.2017.07.017.
[10] 马新娜, 赵猛, 祁琳. 基于卷积脉冲神经网络的故障诊断方法研究[J]. 广西师范大学学报(自然科学版), 2022, 40(3): 112-120. DOI: 10.16088/j.issn.1001-6600.2021070808.
[11] 李辉, 徐伟烝. 噪声干扰下CCSD-CNN轴承故障诊断方法[J/OL]. 轴承, 2022: 1-9[2022-10-26]. http://kns.cnki.net/kcms/detail/41.1148.TH.20220822.1708.002.html.
[12] 张安安, 黄晋英, 冀树伟, 等. 基于卷积神经网络图像分类的轴承故障模式识别[J]. 振动与冲击, 2020, 39(4): 165-171. DOI: 10.13465/j.cnki.jvs.2020.04.021.
[13] 宋向金, 赵文祥. 交流电机信号特征分析的滚动轴承故障诊断方法综述[J]. 中国电机工程学报, 2022, 42(4): 1582-1596. DOI: 10.13334/j.0258-8013.pcsee.210760.
[14] ZHANG W, PENG G L, LI C H. Bearings fault diagnosis based on convolutional neural networks with 2-D representation of vibration signals as input[J]. MATEC Web of Conferences, 2017, 95: 13001. DOI: 10.1051/matecconf/20179513001.
[15] PENG D D, LIU Z L, WANG H, et al. A novel deeper one-dimensional CNN with residual learning for fault diagnosis of wheelset bearings in high-speed trains[J]. IEEE Access, 2019, 7: 10278-10293. DOI: 10.1109/ACCESS.2018.2888842.
[16] 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.
[17] WEN L, LI X, LI X Y, et al. A new transfer learning based on VGG-19 network for fault diagnosis[C]// 2019 IEEE 23rd International Conference on Computer Supported Cooperative Work in Design (CSCWD). Piscataway, NJ: IEEE, 2019: 205-209. DOI: 10.1109/CSCWD.2019.8791884.
[18] HOANG D T, KANG H J. Rolling element bearing fault diagnosis using convolutional neural network and vibration image[J]. Cognitive Systems Research, 2019, 53: 42-50. DOI: 10.1016/j.cogsys.2018.03.002.
[19] 田科位, 董绍江, 姜保军, 等. 基于改进深度残差网络的轴承故障诊断方法[J]. 振动与冲击, 2021, 40(20): 247-254. DOI: 10.13465/j.cnki.jvs.2021.20.031.
[20] 曲建岭, 余路, 袁涛, 等. 基于一维卷积神经网络的滚动轴承自适应故障诊断算法[J]. 仪器仪表学报, 2018, 39(7): 134-143. DOI: 10.19650/j.cnki.cjsi.J1803286.
[21] 曲建岭, 余路, 袁涛, 等. 基于卷积神经网络的层级化智能故障诊断算法[J]. 控制与决策, 2019, 34(12): 2619-2626. DOI: 10.13195/j.kzyjc.2018.0253.
[22] 张训杰, 张敏, 李贤均. 基于二维图像和CNN-BiGRU网络的滚动轴承故障模式识别[J]. 振动与冲击, 2021, 40(23): 194-201, 207. DOI: 10.13465/j.cnki.jvs.2021.23.026.
[23] SANDLER M, HOWARD A, ZHU M L, et al. MobileNetV2: inverted residuals and linear bottlenecks[C]// 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Los Alamitos, CA: IEEE Computer Society, 2018: 4510-4520. DOI: 10.1109/CVPR.2018.00474.
[24] MA N N, ZHANG X Y, ZHENG H T, et al. ShuffleNet V2: practical guidelines for efficient CNN architecture design[C]// Computer Vision -ECCV 2018: LNCS Volume 11218. Cham: Springer, 2018: 122-138. DOI: 10.1007/978-3-030-01264-9_8.
[25] IANDOLA F N, HAN S, MOSKEWICZ M W, et al. SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and <0.5MB model size[EB/OL]. (2016-11-04)[2022-10-26]. https://arxiv.org/abs/1602.07360. DOI: 10.48550/arXiv. 1602.07360.
[26] CHOLLET F. Xception: deep learning with depthwise separable convolutions[C]// 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Los Alamitos, CA: IEEE Computer Society, 2017: 1800-1807. DOI: 10.1109/ CVPR.2017.195.
[27] HU J, SHEN L, SUN G. Squeeze-and-excitation networks[C]// 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Los Alamitos, CA: IEEE Computer Society, 2018: 7132-7141. DOI: 10.1109/CVPR.2018.00745.
[28] RAMACHANDRAN P, ZOPH B, LE Q V. Searching for activation function[EB/OL]. (2017-10-27)[2022-10-26]. https://arxiv.org/abs/1710.05941v2. DOI: 10.48550/arXiv.1710.05941.
[29] 姚齐水, 别帅帅, 余江鸿, 等. 一种结合改进Inception V2模块和CBAM的轴承故障诊断方法[J]. 振动工程学报, 2022, 35(4): 949-957. DOI: 10.16385/j.cnki.issn.1004-4523.2022.04.019.
[30] 陈晓雷, 孙永峰, 李策, 等. 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. DOI: 10.13229/j.cnki.jdxbgxb20211031.
[31] JIN G Q, ZHU T Y, AKRAM M W, et al. An adaptive anti-noise neural network for bearing fault diagnosis under noise and varying load conditions[J]. IEEE Access, 2020, 8: 74793-74807. DOI: 10.1109/ACCESS.2020.2989371.
[32] SHAO S Y, MCALEER S, YAN R Q, et al. Highly accurate machine fault diagnosis using transfer learning[J]. IEEE Transactions on Industrial Informatics, 2019, 15(4): 2446-2455. DOI: 10.1109/TII.2018.2864759.

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed