| Mechanical Engineering |
|
|
|
|
| Wear monitoring of helical milling tool based on one-dimensional convolutional neural network |
Hai-jin WANG( ),Zong-yu YIN,Zhen-zheng KE*(),Ying-jie GUO,Hui-yue DONG |
| Key Laboratory of Advanced Manufacturing Technology of Zhejiang Province, College of Mechanical Engineering, Zhejiang University, Hangzhou 310027, China |
|
|
|
Abstract Helical milling tool wear monitoring methods based on traditional machine learning require complex feature extraction and rich experience, and different wear stages have the same misclassification cost. A new method based on the one-dimensional convolutional neural network (1D CNN) and cost-sensitive learning was introduced, aiming at the above problems and considering the characteristics of one dimension of current signals. Current signals are acquired through spindle, revolution shaft and feed shaft of robotic helical milling end-effector respectively as monitoring signals, and samples are divided using sliding window method to reduce network capacity and increase sample numbers and diversity at the same time. The cost matrix is introduced into the network loss function and the misclassification cost of severe wear stage is increased to make 1D CNN cost sensitive. The time domain current signals are input into 1D CNN directly, and the network can extract tool wear features automatically and unify feature extraction and classification of different wear stages together. Experiment results demonstrate that the tool wear monitoring accuracy of the proposed method is 99.29% and the recall of severe wear stage is 99.60% in the robotic helical milling system.
|
|
Received: 30 November 2019
Published: 05 May 2020
|
|
|
|
Corresponding Authors:
Zhen-zheng KE
E-mail: wanghaijin@zju.edu.cn;kzzcaen@zju.edu.cn
|
基于一维卷积神经网络的螺旋铣刀具磨损监测
基于传统机器学习的螺旋铣刀具磨损监测方法需要复杂的特征提取和丰富的经验知识，不同磨损阶段具有相同的错误分类代价，针对这些问题，结合电流信号一维性特点，提出基于一维卷积神经网络(1D CNN)和代价敏感学习的螺旋铣刀具磨损监测方法. 采集机器人螺旋铣末端执行器主轴、公转轴和进给轴电流作为监测信号，并采用滑动窗口法进行样本划分，在降低网络容量的同时增加样本数量和多样性；在网络损失函数中引入代价矩阵并增加急剧磨损阶段的错误分类代价，使得1D CNN具有代价敏感性；直接将电流时域信号输入1D CNN，网络可以自动提取刀具磨损特征，并将特征提取和不同磨损阶段分类融合在一起. 试验结果表明，在机器人螺旋铣系统中，该方法的刀具磨损监测准确率为99.29%，在急剧磨损阶段的查全率为99.60%.
关键词：
螺旋铣,
刀具磨损监测,
电流信号,
一维卷积神经网络(1D CNN),
代价敏感学习
|
|
| [1] |
PARK K H, BEAL A, KIM D W, et al Tool wear in drilling of composite/titanium stacks using carbide and polycrystalline diamond tools[J]. Wear, 2011, 271 (11/12): 2826- 2835
|
|
|
| [2] |
陈燕, 葛恩德, 傅玉灿, 等 碳纤维增强树脂基复合材料制孔技术研究现状与展望[J]. 复合材料学报, 2015, 32 (2): 301- 316
CHEN Yan, GE En-de, FU Yu-can, et al Review and prospect of drilling technologies for carbon fiber reinforced polymer[J]. Acta Materiae Compositae Sinica, 2015, 32 (2): 301- 316
|
|
|
| [3] |
PEREIRA R B D, BRAND?O L C, PAIVA A P, et al A review of helical milling process[J]. International Journal of Machine Tools and Manufacture, 2017, 120: 27- 48
doi: 10.1016/j.ijmachtools.2017.05.002
|
|
|
| [4] |
KLANCNIK S, FICKO M, BALIC J, et al Computer vision-based approach to end mill tool monitoring[J]. International Journal of Simulation Modelling, 2015, 14 (4): 571- 583
doi: 10.2507/IJSIMM14(4)1.301
|
|
|
| [5] |
GIERLAK P, BURGHARDT A, SZYBICKI D, et al On-line manipulator tool condition monitoring based on vibration analysis[J]. Mechanical Systems and Signal Processing, 2017, 89: 14- 26
doi: 10.1016/j.ymssp.2016.08.002
|
|
|
| [6] |
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
doi: 10.1016/j.ijmachtools.2014.10.011
|
|
|
| [7] |
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
doi: 10.1007/s00170-018-1768-5
|
|
|
| [8] |
LIN X K, ZHOU B, ZHU L Sequential spindle current-based tool condition monitoring with support vector classifier for milling process[J]. The International Journal of Advanced Manufacturing Technology, 2017, 92 (9?12): 3319- 3328
doi: 10.1007/s00170-017-0396-9
|
|
|
| [9] |
PATRA K, JHA A K, SZALAY T, et al Artificial neural network based tool condition monitoring in micro mechanical peck drilling using thrust force signals[J]. Precision Engineering, 2017, 48: 279- 291
doi: 10.1016/j.precisioneng.2016.12.011
|
|
|
| [10] |
沈延斌, 陈岭, 郭浩东, 等 基于深度学习的放置方式和位置无关运动识别[J]. 浙江大学学报: 工学版, 2016, 50 (6): 1141- 1148
SHEN Yan-bin, CHEN Ling, GUO Hao-dong, et al Deep learning based activity recognitionindependent of device orientation and placement[J]. Journal of Zhejiang University: Engineering Science, 2016, 50 (6): 1141- 1148
|
|
|
| [11] |
ZHAO R, YAN R Q, CHEN Z H, et al Deep learning and its applications to machine health monitoring[J]. Mechanical Systems and Signal Processing, 2019, 115: 213- 237
doi: 10.1016/j.ymssp.2018.05.050
|
|
|
| [12] |
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 (9?12): 3217- 3227
doi: 10.1007/s00170-018-2420-0
|
|
|
| [13] |
TERRAZAS G, MARTíNEZ-ARELLANO G, BENARDOS P, et al Online tool wear classification during dry machining using real time cutting force measurements and a CNN approach[J]. Journal of Manufacturing and Materials Processing, 2018, 2 (4): 72
doi: 10.3390/jmmp2040072
|
|
|
| [14] |
KIRANYAZ S, INCE T, ABDELJABER O, et al. 1-D convolutional neural networks for signal processing applications [C]// 2019 IEEE International Conference on Acoustics, Speech and Signal Processing. United Kingdom: IEEE, 2019: 8360-8364.
|
|
|
| [15] |
GIANNAKAKIS G, TRIVIZAKIS E, TSIKNAKIS M, et al. A novel multi-kernel 1D convolutional neural network for stress recognition from ECG [C]// International Conference on Affective Computing and Intelligent Interaction Workshops and Demos. United Kingdom: ACⅡW, 2019: 1-4.
|
|
|
| [16] |
ZHANG Y Q, MIYAMORI Y, MIKAMI S C, et al Vibration-based structural state identification by a 1-dimensional convolutional neural network[J]. Computer-Aided Civil and Infrastructure Engineering, 2019, 34 (9): 822- 839
doi: 10.1111/mice.12447
|
|
|
| [17] |
WU C Z, JIANG P C, DING C, et al Intelligent fault diagnosis of rotating machinery based on one-dimensional convolutional neural network[J]. Computers in Industry, 2019, 108: 53- 61
doi: 10.1016/j.compind.2018.12.001
|
|
|
| [18] |
ZHANG C, TAN K C A cost-sensitive deep belief network for imbalanced classification[J]. IEEE Transactions on Neural Networks and Learning Systems, 2019, 30 (1): 109- 122
doi: 10.1109/TNNLS.2018.2832648
|
|
|
| [19] |
XU G, ZHOU H, CHEN J CNC internal data based incremental cost-sensitive support vector machinemethod for tool breakage monitoring in end milling[J]. Engineering Applications of Artificial Intelligence, 2018, 74: 90- 103
doi: 10.1016/j.engappai.2018.05.007
|
|
|
| [20] |
董辉跃, 朱灵盛, 章明, 等 飞机蒙皮切边的螺旋铣削[J]. 浙江大学学报:工学版, 2015, 49 (11): 2033- 2039
DONG Hui-yue, ZHU Ling-sheng, ZHANG Ming, et al Orbital milling method of aircraft skins trimming[J]. Journal of Zhejiang University: Engineering Science, 2015, 49 (11): 2033- 2039
|
|
|
| [21] |
李宏坤, 郝佰田, 代月帮, 等 基于压缩感知和加噪堆栈稀疏自编码器的铣刀磨损程度识别方法研究[J]. 机械工程学报, 2019, 55 (14): 1- 10
LI Hong-kun, HAO Bai-tian, DAI Yue-bang, et al Wear status recognition for milling cutter based on compressed sensing and noise stacking sparse auto-encoder[J]. Journal of Mechanical Engineering, 2019, 55 (14): 1- 10
|
|
|
| [22] |
KHAN S H, HAYAT M, BENNAMOUN M, et al Cost-sensitive learning of deep feature representations from imbalanced data[J]. IEEE Transactions on Neural Networks and Learning Systems, 2017, 29 (8): 3573- 3587
|
|
|
|
Viewed |
|
|
|
Full text
|
|
|
|
|
Abstract
|
|
|
|
|
Cited |
|
|
|
|
| |
Shared |
|
|
|
|
| |
Discussed |
|
|
|
|
