基于域对抗门控网络的变工况刀具磨损精确预测方法

万鹏; 李迎光; 刘长青; 华家玘

doi:10.7527/S1000-6893.2020.24879

航空学报 >

2021 , Vol. 42 >Issue 10: 524879 - 524879

DOI: https://doi.org/10.7527/S1000-6893.2020.24879

论文

基于域对抗门控网络的变工况刀具磨损精确预测方法

万鹏 ,
李迎光 ,
刘长青 ,
华家玘

展开

南京航空航天大学机电学院, 南京 210016

收稿日期: 2020-10-15

修回日期: 2020-11-03

网络出版日期: 2020-12-08

基金资助

国家重点研发计划（2018YFB1703203）；国家自然科学基金杰出青年基金（51925505）；国家自然科学基金（51775278）

收起

Method for accurate prediction of tool wear under varying cutting conditions based on domain adversarial gating neural network

WAN Peng ,
LI Yingguang ,
LIU Changqing ,
HUA Jiaqi

Expand

College of Mechanical and Electrical Engineering, Nanjing University of Aeronautics & Astronautics, Nanjing 210016, China

Received date: 2020-10-15

Revised date: 2020-11-03

Online published: 2020-12-08

Supported by

National Key Research and Development Program of China (2018YFB1703203); National Science Foundation of China for Distinguished Young Scholars (51925505); National Natural Science Foundation of China (51775278)

Fold

摘要

刀具磨损的精确预测对保证零件加工质量、提高生产效率和降低制造成本具有重要作用。在实际加工过程中，切削参数、刀具几何参数、刀具材料等工况复杂多变，工况信息和刀具磨损量对监测信号的耦合作用为刀具磨损的精确预测带来了很大挑战。针对以上问题，提出了一种基于域对抗门控网络（DAGNN）的变工况刀具磨损精确预测方法。引入工况分类网络并利用无磨损量标签样本，通过域对抗和门控过滤机制自适应地从不同工况的原始监测信号中提取表征刀具磨损且对工况变化不敏感的关键信号特征。对信号特征提取网络和刀具磨损预测网络进行迭代优化，从而实现变工况刀具磨损的精确预测。实验结果表明：相比已有的方法，本文方法能够利用少量带磨损量标签的目标工况样本实现刀具材料和刀具直径变化情况下的刀具磨损量精确预测，预测精度大幅提高。

关键词： 刀具磨损预测; 变工况; 特征提取; 域对抗门控网络; 小样本学习

本文引用格式

万鹏 , 李迎光 , 刘长青 , 华家玘 . 基于域对抗门控网络的变工况刀具磨损精确预测方法[J]. 航空学报, 2021 , 42(10) : 524879 -524879 . DOI: 10.7527/S1000-6893.2020.24879

Abstract

The accurate prediction of tool wear plays an important role in ensuring the quality of parts, improving production efficiency and reducing manufacturing costs. In the actual machining process, cutting parameters, tool geometric parameters, tool materials and other cutting conditions are complex and variable. The coupling effect of cutting condition and tool wear on monitoring signals brings great challenges to accurate prediction of tool wear. To address the above issue, this paper proposes a method for accurate prediction of tool wear under variable cutting conditions based on Domain Adversarial Gating Neural Network (DAGNN). A cutting condition classification network is trained using unlabeled samples, and key signal features, which can represent tool wear and are not sensitive to cutting conditions, are adaptively extracted from the original monitoring signals of different cutting conditions through domain adversarial and gated filtering mechanisms. The signal feature extraction network and tool wear prediction network are iteratively optimized to achieve accurate prediction of tool wear under variable cutting conditions. Experimental results show that compared with existing methods, the method proposed can achieve higher prediction accuracy even when only a few wear labeled samples of target cutting condition are available.

Key words： tool wear prediction; varying cutting condition; feature extraction; domain adversarial gating neural network; few-shot learning

参考文献

[1] 刘具龙, 张璧, 白倩, 等. 钛合金铣削刀具/工件接触区域温度预测[J]. 航空学报, 2018, 39(12):422128. LIU J L, ZHANG B, BAI Q, et al. Temperature prediction of tool/workpiece contact zone in titanium milling[J]. Acta Aeronautica et Astronautica Sinica, 2018, 39(12):422128(in Chinese).
[2] 陈燕, 杨树宝, 傅玉灿, 等. 钛合金TC4高速切削刀具磨损的有限元仿真[J]. 航空学报, 2013, 34(9):2230-2240. CHEN Y, YANG S B, FU Y C, et al. FEM estimation of tool wear in high speed cutting of Ti₆Al₄V alloy[J]. Acta Aeronautica et Astronautica Sinica, 2013, 34(9):2230-2240(in Chinese).
[3] KONG D D, CHEN Y J, LI N, et al. Tool wear monitoring based on kernel principal component analysis and v-support vector regression[J]. The International Journal of Advanced Manufacturing Technology, 2017, 89(1-4):175-190.
[4] 王强. 基于深度学习的数控加工刀具寿命动态评估方法[D]. 南京:南京航空航天大学, 2019. WANG Q. Dynamic evaluation method of tool life in NC machining based on deep learning[D]. Nanjing:Nanjing University of Aeronautics & Astronautics, 2019(in Chinese).
[5] SALONITIS K, KOLIOS A. Reliability assessment of cutting tool life based on surrogate approximation methods[J]. The International Journal of Advanced Manufacturing Technology, 2014, 71(5-8):1197-1208.
[6] 倪金成. 数控加工刀具多失效形式状态监测[D]. 南京:南京航空航天大学, 2020. NI J C. Monitoring of Multiple Failure Forms of NC Cutting Tools[D]. Nanjing:Nanjing University of Aeronautics & Astronautics, 2020(in Chinese).
[7] 隋少春, 许艾明, 黎小华, 等. 面向航空智能制造的DT与AI融合应用[J]. 航空学报, 2020, 41(7):624173. SUI S C, XU A M, LI X H, et al. Fusion application of DT and AI for aviation intelligent manufacturing[J]. Acta Aeronautica et Astronautica Sinica, 2020, 41(7):624173(in Chinese).
[8] TAO F, QI Q L, LIU A, et al. Data-driven smart manufacturing[J]. Journal of Manufacturing Systems, 2018, 48:157-169.
[9] ZHOU Y Q, XUE W. Review of tool condition monitoring methods in milling processes[J]. The International Journal of Advanced Manufacturing Technology, 2018, 96:2509-2523.
[10] WANG G F, YANG Y W, XIE Q L, et al. Force based tool wear monitoring system for milling process based on relevance vector machine[J]. Advances in Engineering Software, 2014, 71:46-51.
[11] 刘宇, 王迁, 刘阔, 等. 基于小波奇异性和支持向量机微铣刀破损检测[J]. 东北大学学报(自然科学版), 2017, 38(10):1426-1430. LIU Y, WANG Q, LIU K, et al. Micro milling cutter breakage detection based on wavelet singularity and support vector machine[J]. Journal of Northeastern University (Natural Science), 2017, 38(10):1426-1430(in Chinese).
[12] CHEN J Q, CHEN H B, XU J J, et al. Acoustic signal-based tool condition monitoring in belt grinding of nickel-based superalloys using RF classifier and MLR algorithm[J]. The International Journal of Advanced Manufacturing Technology, 2018, 98(1-4):859-872.
[13] CORNE R, NATH C, EL MANSORI M, et al. Study of spindle power data with neural network for predicting real-time tool wear/breakage during inconel drilling[J]. Journal of Manufacturing Systems, 2017, 43:287-295.
[14] 牟文平. 数据驱动的数控加工刀具磨损量精确预测关键技术研究[D]. 南京:南京航空航天大学, 2020:9-11. MOU W P. Key technologies on data-driven real time prediction of tool wear for NC machining[D]. Nanjing:Nanjing University of Aeronautics &Astronautics, 2020:9-11(in Chinese).
[15] WANG J J, MA Y L, ZHANG L B, et al. Deep learning for smart manufacturing:Methods and applications[J]. Journal of Manufacturing Systems, 2018, 48:144-156.
[16] 林杨, 高思煜, 刘同舜, 等. 基于深度学习的高速铣削刀具磨损状态预测方法[J]. 机械与电子, 2017, 35(7):12-17. LIN Y, GAO S Y, LIU T S, et al. A deep learning-based method for tool wear state prediction in high speed milling[J]. Machinery & Electronics, 2017, 35(7):12-17(in Chinese).
[17] FU Y, ZHANG Y, QIAO H Y, et al. Analysis of feature extracting ability for cutting state monitoring using deep belief networks[J]. Procedia CIRP, 2015, 31:29-34.
[18] HUANG Z W, ZHU J M, LEI J T, et al. Tool wear predicting based on multisensory raw signals fusion by reshaped time series convolutional neural network in manufacturing[J]. IEEE Access, 2019, 7:178640-178651.
[19] CUKA B, KIM D W. Fuzzy logic based tool condition monitoring for end-milling[J]. Robotics and Computer-Integrated Manufacturing, 2017, 47:22-36.
[20] SHI X H, WANG R, CHEN Q T, et al. Cutting sound signal processing for tool breakage detection in face milling based on empirical mode decomposition and independent component analysis[J]. Journal of Vibration and Control, 2015, 21(16):3348-3358.
[21] HONG Y S, YOON H S, MOON J S, et al. Tool-wear monitoring during micro-end milling using wavelet packet transform and Fisher's linear discriminant[J]. International Journal of Precision Engineering and Manufacturing, 2016, 17(7):845-855.
[22] KALVODA T, HWANG Y R. A cutter tool monitoring in machining process using Hilbert-Huang transform[J]. International Journal of Machine Tools and Manufacture, 2010, 50(5):495-501.
[23] NOURI M, FUSSELL B K, ZINITI B L, et al. Real-time tool wear monitoring in milling using a cutting condition independent method[J]. International Journal of Machine Tools and Manufacture, 2015, 89:1-13.
[24] 刘宇, 汪惠芬, 刘庭煜. 一种基于多特征和支持向量机的刀具磨损状态识别技术[J]. 制造业自动化, 2016, 38(5):132-138. LIU Y, WANG H F, LIU T Y. An intelligent tool wear estimation technology based on multi-feature and support vector machine[J]. Manufacturing Automation, 2016, 38(5):132-138(in Chinese).
[25] LI Y G, LIU C Q, HUA J Q, et al. A novel method for accurately monitoring and predicting tool wear under varying cutting conditions based on meta-learning[J]. CIRP Annals, 2019, 68(1):487-490.
[26] GANIN Y, LEMPITSKY V. Unsupervised domain adaptation by backpropagation[DB/OL]. arXiv preprint:1409.7495, 2014.
[27] WANG Z R, DAI Z H, PÓCZOS B, et al. Characterizing and avoiding negative transfer[C]//2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Piscataway:IEEE Press, 2019:11285-11294.
[28] CHENG C, LI J Y, LIU Y M, et al. Deep convolutional neural network-based in-process tool condition monitoring in abrasive belt grinding[J]. Computers in Industry, 2019, 106:1-13.
[29] AGHAZADEH F, TAHAN A, THOMAS M. Tool condition monitoring using spectral subtraction and convolutional neural networks in milling process[J]. The International Journal of Advanced Manufacturing Technology, 2018, 98:3217-3227.
[30] ZHANG H X, ZHAO J, WANG F Z, et al. Cutting forces and tool failure in high-speed milling of titanium alloy TC21 with coated carbide tools[J]. Proceedings of the Institution of Mechanical Engineers Part B:Journal of Engineering Manufacture, 2015, 229(1):20-27.
[31] GOODFELLOW I J, POUGET-ABADIE J, MIRZA M, et al. Generative adversarial nets[DB/OL]. arXiv preprint:1406.2661, 2014.
[32] 李晶晶, 李迎光, 刘长青, 等. 数控加工过程多源数据实时采集与同步方法[J]. 工具技术, 2019, 53(7):106-110. LI J J, LI Y G, LIU C Q, et al. Real-time acquisition and synchronization method of multi-source data in CNC machining process[J]. Tool Engineering, 2019, 53(7):106-110(in Chinese).
[33] International Organization for Standardization. Tool life testing in milling-Part 2:End milling:ISO 8688-2[S]. Geneva:International Organization for Standardization, 1989.

Options

文章导航

模态框（Modal）标题

摘要

本文引用格式

Abstract

参考文献