基于S-L2UWE变换和SA-ConvNeXt的永磁同步直线电机导轨故障诊断

PDF(5770 KB)

振动与冲击 ›› 2024, Vol. 43 ›› Issue (19) : 28-36.

论文

基于S-L2UWE变换和SA-ConvNeXt的永磁同步直线电机导轨故障诊断

吴艳萍1，陆思良1，宋俊材2，王骁贤3，吴先红1，丁伟1

作者信息 +

Fault diagnosis of permanent magnet synchronous linear motor guide railbased on S-L2UWE transform and SA-ConvNeXt

WU Yanping1, LU Siliang1, SONG Juncai2, WANG Xiaoxian3, WU Xianhong1, DING Wei1

Author information +

文章历史 +

摘要

本文提出了一种基于S-L2UWE（SL）变换和SA-ConvNeXt模型相结合的故障诊断方法，以解决永磁同步直线电机导轨磨损、滚珠磨损和滚珠缺失等多类型故障精准识别诊断的问题。首先对采集的振动信号进行S变换处理，将一维时域信号转化成二维时频图像；利用L2UWE算法对S变换后的图像进行增强，实现故障时频图像颜色和形状的细节特征增强显示；将自注意力模块（self-attention，SA）引入到ConvNeXt模型中，从特征通道层面上统计图像的全局信息，使网络关注重点特征，提高模型分类精度。最终，基于SL变换和SA-ConvNeXt深度学习模型对于直线导轨故障的识别准确率高达93.8%，能准确识别导轨和滚珠磨损、缺失等故障。对比实验和鲁棒性实验，验证了本文所提方法的有效性和优越性。

Abstract

In this paper, a fault diagnosis method based on S-L2UWE (SL) transform and SA-ConvNeXt model is proposed to solve the problem of accurate identification and diagnosis of multiple types of faults such as guide wear, ball wear and ball loss of permanent magnet synchronous linear motor. Firstly, the collected vibration signal is processed by S-transform, and the one-dimensional time-domain signal is converted into two-dimensional time-frequency image. The L2UWE algorithm was used to enhance the S-transformed image to realize the enhanced display of the fault video image details of color and shape. The self-attention module (SA) is introduced into the ConvNeXt model to calculate the global information of the image from the feature channel level, so that the network focuses on the key features and improves the classification accuracy of the model. Finally, the fault recognition accuracy of linear guide rail faults based on SL transform and SA-ConvNeXt deep learning model is as high as 93.8%, which can accurately identify faults such as guide rail and ball wear and loss. Comparative experiments and robustness experiment verify the effectiveness and superiority of the proposed method.

导出引用

吴艳萍1, 陆思良1, 宋俊材2, 王骁贤3, 吴先红1, 丁伟1. 基于S-L2UWE变换和SA-ConvNeXt的永磁同步直线电机导轨故障诊断[J]. 振动与冲击, 2024, 43(19): 28-36

WU Yanping1, LU Siliang1, SONG Juncai2, WANG Xiaoxian3, WU Xianhong1, DING Wei1. Fault diagnosis of permanent magnet synchronous linear motor guide railbased on S-L2UWE transform and SA-ConvNeXt[J]. Journal of Vibration and Shock, 2024, 43(19): 28-36

参考文献

[1] CUI F, SUN Z, XU W, et al. Optimization Analysis of Long Primary Permanent Magnet Linear Synchronous Motor [J]. IEEE Transactions on Applied Superconductivity, 2021, 31(8): 1-4.
[2] LI J, HUANG X, ZHOU B, et al. Design Criterion for Dual-Modular Permanent Magnet Linear Synchronous Motor With Long Stroke and High Flexibility [J]. IEEE Transactions on Industrial Electronics, 2022, 69(10): 10428-10436.
[3] MA M, TAO W, ZHANG X, et al. Electromagnetic Interference Analysis of Permanent Magnet Linear Synchronous Launchers [J]. IEEE Transactions on Plasma Science, 2020, 48(5): 1309-1315.
[4] 董家臣, 高钦和, 陈志翔. 基于EKF-LESO级联观测的永磁直线同步电机无传感器线性自抗扰控制 [J]. 振动与冲击, 2019, 38(21): 53-61+70.
DONG Jiachen, GAO Qinhe, CHEN Zhixiang. Sensorless LADRC System of PMLSM based on EKF-LESO cascade observation. JOURNAL OF VIBRATION AND SHOCK, 2019, 38(21): 53-61+70.
[5] 刘铄, 宋俊材, 陆思良. 基于灰度纹理特征提取和CS-SNN的双初级永磁同步直线电机退磁故障诊断研究 [J]. 中国电机工程学报, 2023, 43(16): 6464-6474.
LIU Shuo, SONG Juncai, LU Siliang. Demagnetization Fault Diagnosis Research of DPPMSLM Based on Gray Texture Feature Extraction and CS-SNN.Proceedings of the CSEE. 2023, 43(16): 6464-6473.
[6] BIANCHINI C, IMMOVILLI F, COCCONCELLI M, et al. Fault Detection of Linear Bearings in Brushless AC Linear Motors by Vibration Analysis [J]. IEEE Transactions on Industrial Electronics, 2011, 58(5): 1684-1694.
[7] KIM M S, YUN J P, PARK P. An Explainable Convolutional Neural Network for Fault Diagnosis in Linear Motion Guide [J]. IEEE Transactions on Industrial Informatics, 2021, 17(6): 4036-4045.
[8] JIAN H, LEE H-R, AHN J-H. Detection of Bearing/Rail Defects for Linear Motion Stage Using Acoustic Emission [J]. International Journal of Precision Engineering and Manufacturing, 2013, 14(11): 2043-2046.
[9] PRATHAN C, THANASAK W, JIRAPHON S. Fault detection and diagnosis of linear bearing in auto core adhesion mounting machines based on condition monitoring [J]. Systems Science & Control Engineering, 2021, 9(1): 290-303.
[10] SIWANU L, PRATHAN C, JIRAPHON S. Linear Bearing Fault Detection in Operational Condition Using Artificial Neural Network [J]. ITM Web of Conferences, 2019, 24:01004.
[11] 乔志城, 刘永强, 廖英英. 改进经验小波变换与最小熵解卷积在铁路轴承故障诊断中的应用 [J]. 振动与冲击, 2021, 40(02): 81-90.
QIAO Zhicheng，LIU Yongqiang，LIAO Yingying. Application of improved wavelet transform and minimum entropy deconvolution in railway bearing fault diagnosis. Journal of Vibration and Shock, 2021, 40(2): 81-90.
[12] 黄斯琪, 谭志银, 杨思国. 改进的傅里叶分解方法及其在滚动轴承故障诊断中的应用 [J]. 振动与冲击, 2023, 42(12): 178-186.
HUANG Siqi,TAN Zhiyin,YANG Siguo. An improved Fourier decomposition method and its application in fault diagnosis of rolling bearings. Journal of Vibration and Shock, 2023, 42(12): 178-186.
[13] 昝涛, 王辉, 刘智豪. 基于多输入层卷积神经网络的滚动轴承故障诊断模型 [J]. 振动与冲击, 2020, 39(12): 142-149+163.
ZAN Tao, WANG Hui, LIU Zhihao. A fault diagnosis model for rolling bearings based on a multi-input layer convolutional neural network. Journal of Vibration and Shock, 2020, 39(12): 142-149+163.
[14] FENG H, CHEN R, WANG Y. Feature extraction for fault diagnosis based on wavelet packet decomposition: An application on linear rolling guide [J]. Advances in Mechanical Engineering, 2018, 10(8): 1687814018796367.
[15] HONG D, BANG S, KIM B. Unsupervised Condition Diagnosis of Linear Motion Guide Using Generative Model Based on Images [J]. IEEE Access, 2021, 9: 80491-80499.
[16] BODUROGLU A, DEMIR Y, CUMHUR B, et al. A Novel Track Structure of Double-Sided Linear PM Synchronous Motor for Low Cost and High Force Density Applications [J]. IEEE Transactions on Magnetics, 2021, 57(2): 1-5.
[17] TAO Y, CAO S, MA Y, et al. Second-Order Adaptive Synchrosqueezing S Transform and Its Application in Seismic Ground Roll Attenuation [J]. IEEE Geoscience and Remote Sensing Letters, 2020, 17(8): 1308-1312.
[18] WANG Q, LI Y, CHEN S, et al. Matching Demodulation Synchrosqueezing S Transform and Its Application in Seismic Time-Frequency Analysis [J]. IEEE Geoscience and Remote Sensing Letters, 2022, 19: 1-5.
[19] XU Y, WANG L, HU A, et al. Time-extracting S-transform algorithm and its application in rolling bearing fault diagnosis [J]. Science China-Technological Sciences, 2022, 65(4): 932-942.
[20] 杜小磊, 陈志刚, 王衍学. 基于同步挤压S变换和集成深层脊波自编码器的轴承故障诊断 [J]. 振动与冲击, 2020, 39(14): 59-68.
DU Xiaolei，CHEN Zhigang，WANG Yanxue. Bearing fault diagnosis based on the synchrosqueezed S transform and ensemble deep ridgelet auto-encoder. JOURNAL OF VIBRATION AND SHOCK, 2020, 39(14): 59-68.
[21] MARQUES T P, ALBU A B. L2UWE: A Framework for the Efficient Enhancement of Low-Light Underwater Images Using Local Contrast and Multi-Scale Fusion [C] // proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops. 2020.
[22] HUANG Z, LI J, HUA Z, et al. Underwater Image Enhancement via Adaptive Group Attention-Based Multiscale Cascade Transformer [J]. IEEE Transactions on Instrumentation and Measurement, 2022, 71: 1-18.
[23] ZHU Y, SUN X, ZHANG H, et al. Near-infrared and visible fusion for image enhancement based on multi-scale decomposition with rolling WLSF [J]. Infrared Physics & Technology, 2023, 128: 104434.
[24] LIU Z, MAO H, WU C Y, et al. A ConvNet for the 2020s [C]// proceedings of the 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). 2022 .
[25] ZHANG Y, WANG H, YANG G, et al. CSNet: a ConvNeXt-based Siamese network for RGB-D salient object detection [J]. The Visual Computer, 2023: 1-19.
[26] TANG X, LI B, GUO J, et al. A Cross-Modal Feature Fusion Model Based on ConvNeXt for RGB-D Semantic Segmentation [J]. Mathematics, 2023, 11(8): 1828.
[27] ZHANG C, QIN F, ZHAO W, et al. Research on Rolling Bearing Fault Diagnosis Based on Digital Twin Data and Improved ConvNext [J]. Sensors, 2023, 23(11): 5534.
[28] LI L, HAN L, CAO H, et al. Joint Self-Attention for Remote Sensing Image Matching [J]. IEEE Geoscience and Remote Sensing Letters, 2022, 19: 1-5.