基于空间坐标与振动特征融合的机床切削状态分类方法

摘要
图/表
参考文献(26)
相关文章 (15)

全文: PDF (2364 KB) HTML (1 KB)
输出: BibTeX | EndNote (RIS)

摘要 t分布的随机邻域嵌入（t-distributed stochastic neighbor embedding, t-SNE）常被用作机床切削状态分类中的特征选择方法，以学习切削参数之间的潜在关系。为了提高切削状态分类的精度，融合振动信号特征与切削激励点的空间坐标，提出了空间坐标嵌入的t分布的随机邻域嵌入方法（spatial coordinate embedded t-SNE, Ct-SNE）。该方法采用振动信号构建高维特征空间，将空间坐标作为物理信息嵌入至特征空间，以优选出类内相似度高、类间差异性大的特征。实验采集了三轴立式铣床加工的数据，对比了传统t-SNE方法与Ct-SNE方法的可视化结果和切削状态分类的准确性。结果表明，与传统方法相比，切削激励点的空间坐标的引入可以提高振动特征的可区分度，显著提升切削状态分类的准确率。
关键词：状态监测；t分布的随机邻域嵌入；特征选择；振动监测；空间坐标

	服务

	把本文推荐给朋友
	加入我的书架
	加入引用管理器
	E-mail Alert
	RSS
	作者相关文章
	王晶1
	2
	程晓斌1
	2
	高艳1
	王勋1
	杨军1
	2

关键词 ：状态监测, t分布的随机邻域嵌入, 特征选择, 振动监测, 空间坐标

Abstract：The t-distributed stochastic neighbor embedding (t-SNE) has been used as a feature selection method in condition classification to learn potential relationship between cutting parameters. To improve the accuracy of cutting condition classification, the spatial coordinate embedded t-SNE method (Ct-SNE) is proposed by fusing the vibration signals and spatial coordinates of cutting excitation point. In this method, vibration signals are used to construct high-dimensional features space, with spatial coordinates being embedded as physical information for feature selection with high intra-class similarity and high inter-class variability. Experiments were carried out on a three-axis vertical milling machine to collect cutting data. The visualization results and cutting condition classification results of Ct-SNE were compared with the traditional method t-SNE. The experimental results show that the Ct-SNE method can improve the distinguishability of vibration features and significantly increase the accuracy of cutting condition classification.
Key words: condition monitoring; t-SNE; feature selection; vibration monitoring; spatial coordinate

Key words： condition monitoring t-SNE feature selection vibration monitoring spatial coordinate

收稿日期: 2021-10-14 出版日期: 2022-12-15

引用本文:

王晶1,2，程晓斌1,2，高艳1，王勋1，杨军1,2. 基于空间坐标与振动特征融合的机床切削状态分类方法[J]. 振动与冲击, 2022, 41(23): 257-264.
WANG Jing1,2, CHENG Xiaobin1,2, GAO Yan1, WANG Xun1, YANG Jun1,2. Machine tool cutting state classification method based on fusion of spatial coordinate and vibration features. JOURNAL OF VIBRATION AND SHOCK, 2022, 41(23): 257-264.

链接本文:

http://jvs.sjtu.edu.cn/CN/ 或 http://jvs.sjtu.edu.cn/CN/Y2022/V41/I23/257

[1] DEEPAM G, PABLA B S. Condition based maintenance of machine tools—A review [J]. Cirp Journal of Manufacturing Science and Technology, 2015, 10:24-35.
[2] 贾志成，胡仲翔，申桂香，王桂萍. 数控机床故障模式的对比分析[J]. 兵工学报，2009, 30(1):86-90.
JIA Zhicheng, HU Zhongxiang, SHEN Guixiang, WANG Guiping. Contrast of fault mode analyses of CNC machine tools between the Chinese- built and the German [J]. Acta Armamentarii, 2009, 30(1):86-90.
[3] 李梦群. 现代数控机床故障诊断及维修第4版[M]. 北京：国防工业出版社，2016.
[4] 王丰，陈建波，周进，周文昌，谢睿. 基于柔性生产线的数控加工过程研究[J]. 组合机床与自动化加工技术，2021, 8:5.
WANG Feng, CHEN Jianbo, ZHOU Jin, XIE Rui. Research on NC machining process based on flexible production line [J]. Modular Machine Tool & Automatic Manufacturing Technique, 2021, 8:5.
[5] LAURO C H, BRANDÃO L C, BALDO D, et al. Monitoring and processing signal applied in machining processes-A review [J]. Measurement, 2014, 58:73-86.
[6] MONTAZERI A, ANSARIZADEH M H, AREFI M M. A data-driven statistical approach for monitoring and analysis of large industrial processes [J]. IFAC-PapersOnLine, 2019, 52(13):2354-2359.
[7] 安煌，赵荣珍. 转子故障数据集降维的CKLPMDP算法研究[J]. 振动与冲击，2021, 40(9):37-42.
AN Huang, ZHAO Rongzhen. CKLPMDP algorithm for dimension reduction of a rotor fault data set [J]. Journal of Vibration and Shock, 2021, 40(9):37-42.
[8] WU X, ZHANG Y, CHENG C, et al. A hybrid classification autoencoder for semi-supervised fault diagnosis in rotating machinery [J]. Mechanical Systems and Signal Processing, 2021, 149:107327.
[9] TETI R, JEMIELNIAK K, G. O’DONNELL, et al. Advanced monitoring of machining operations [J]. CIRP Annals - Manufacturing Technology, 2010, 59(2):717-739.
[10] 王振亚，姚立纲，蔡永武，张俊. 基于熵-流特征和樽海鞘群优化支持向量机的故障诊断方法[J]. 振动与冲击, 2021, 40(6):107-114.
WANG Zhenya，YAO Ligang，CAI Yongwu，ZHANG Jun. Fault diagnosis method based on the entropy-manifold feature and SSO-SVM [J]. Journal of Vibration and Shock, 2021, 40(6):107-114.
[11] 张绍辉. 基于流形学习的机械状态识别方法研究[D]. 广州：华南理工大学, 2014.
[12] DER MAATEN L V, HINTON G E. Visualizing data using t-SNE [J]. Journal of Machine Learning Research, 2008, 2579-2605.
[13] ZHENG J, JIANG Z, Pan H. Sigmoid-based refined composite multiscale fuzzy entropy and t-SNE based fault diagnosis approach for rolling bearing [J]. Measurement, 2018, 129:332-342.
[14] YU W, WU M, LU C, Meticulous process monitoring with multiscale convolutional feature extraction [J]. Journal of Process Control, 2021, 106:20-28.
[15] KERMAN L, SUSANA F, AITOR A, BASILIO S. Comparison of automated feature selection and reduction methods on the condition monitoring issue [J]. Procedia Manufacturing, 2018, 16:2-9.
[16] TANG Q, CHAI Y, QU J, FANG Xi. Industrial process monitoring based on Fisher discriminant global-local preserving projection [J]. Journal of Process Control, 2019, 81:76-86.
[17] 郑建炜, 邱虹, 蒋一波,等. 判别随机近邻嵌入分析方法[J]. 计算机辅助设计与图形学学报, 2012, 024(011):1477-1484.
ZHENG Jianyi, QIU Hong, JIANG Yibo, et al. Discriminative Stochastic Neighbor Embedding Analysis Method [J]. Journal of Computer-Aided Design & Computer Graphics, 2012, 024(011):1477-1484.
[18] CHENG J, LIU H, WANG F, et al. Silhouette analysis for human action recognition based on supervised temporal t-SNE and incremental learning [J]. IEEE Transactions on Image Processing, 2015, 24(10):3203-3217.
[19] YU M, ZHANG S, ZHAO L, et al. Deep supervised t-SNE for SAR target recognition [C]. 2017 2nd International Conference on Frontiers of Sensors Technologies (ICFST), 2017, 265-269.
[20] 姚嵩，胡于进，王学林. 工件几何模型对不锈钢切削力数值计算影响研究[J]. 中国机械工程，2011, 22(12):1392-1396.
YAO Song, HU Yujin, WANG Xuelin. Influence of Workpiece Geometry on Numerical Calculation of Cutting Force for Stainless Steel [J]. China Mechanical Engineering, 2011, 22(12):1392-1396.
[21] 吴卫国. 高效精密切削及其振动特性的研究[D]. 镇江：江苏大学, 2007.
[22] CHENG J, LIU H, LI H. Silhouette analysis for human action recognition based on maximum spatio-temporal dissimilarity embedding [J]. Machine Vision and Application, 2014, 25(4):1007–1018.
[23] ELBHBAH K, SINHA J K. Vibration-based condition monitoring of rotating machines using a machine composite spectrum [J]. Journal of Sound and Vibration, 2013, 332(11):2831-2845.
[24] JACOBS R A. Increased rates of convergence through learning rate adaptation [J]. Neural Networks, 1988, 1(4):295-307.
[25] XIANG S M, NIE F P, SONG Y Q, et al. Embedding new date for manifold learning via coordinate propagation [J]. Knowledge and Information Systems Journal, 2009, 19:159-184.
[26] SALTON G, MCGILL M. Introduction to modern information retrieval [M]. New York: McGraw-Hill, Inc., 1986.