Tool Wear Condition Recognition Based on SDAE

Abstract

Abstract: A new method of condition recognition for tool wear was proposed based on SADE . A SDAE neural network was constructed to learn the characteristics of AE signals, and a supervised fine-tuning of the autoencoder network was carried out, so that the tool wear conditions were accurately recognized. The experimental results show that the SDAE method may learn adaptively to get effective feature expressions and the tool wear condition recognition precision is high. The proposed method may be used to recognize tool wear conditions effectively.

Key words: tool wear, acoustic emission（AE）, deep learning, stacked denoising autoencoder （SDAE）

摘要： 提出了一种基于堆叠降噪自编码（SDAE）的刀具磨损状态识别方法。构建了SDAE神经网络来学习声发射（AE）信号的特征,并对自编码网络进行有监督的微调,从而对刀具磨损状态进行精确识别。实验结果表明,SDAE方法能够自适应地学习,得到有效的特征表达,且刀具磨损状态识别结果精确度高,该方法能够有效地进行刀具磨损状态识别。

关键词: 刀具磨损, 声发射, 深度学习, 堆叠降噪自编码

CLC Number:

TG502

WANG Lihua1;YANG Jiawei1;ZHANG Yonghong1;ZHAO Xiaoping2;XIE Yangyang1. Tool Wear Condition Recognition Based on SDAE[J]. China Mechanical Engineering.

王丽华1;杨家巍1;张永宏1;赵晓平2;谢阳阳1. 基于堆叠降噪自编码的刀具磨损状态识别[J]. 中国机械工程.

References

[1]王丽华,陶润喆,张永宏,等.基于CEEMD-WPT的滚动轴承特征提取算法[J].振动、测试与诊断,2017,37（1）：181-188.
WANG Lihua, TAO Runzhe, ZHANG Yonghong, et al. Feature Extraction of Rolling Bearing Based on CEEMD-WPT[J]. Journal of Vibration,Measurement & Diagnosis, 2017,37（1）：181-188.
[2]陶润喆. 基于CEEMD-WPT的刀具磨损状态识别研究[D].南京：南京信息工程大学, 2017.
TAO Runzhe. Research on State Recognition of Tool Wear Based on CEEWD-WPT[D]. Nanjing： Nanjing University of Information Science & Technology, 2017.
[3]谢阳阳. 基于声发射和深度学习的刀具状态识別研究[D].南京：南京信息工程大学, 2017.
XIE Yangyang. Research on Tool State Recognition Based on Acoustic Emission and Deep Learning[D]. Nanjing： Nanjing University of Information Science & Technology, 2017.
[4]HINTON G E, SIMON O, TEH Y W. A Fast Learning Algorithm for Deep Belief Nets[J]. Neural computation, 2006, 18（7）：1527-1554.
[5]王丽华,谢阳阳,张永宏,等. 采用深度学习的异步电机故障诊断方法[J].西安交通大学学报, 2017,51（10）：128-134.
WANG Lihua, XIE Yangyang, ZHANG Yonghong, et al. A Fault Diagnosis Method for Asynchronous Motor Using Deep Learning[J]. Journal of Xi'an Jiaotong University,2017,51（10）：128-134.
[6]雷亚国,贾峰,周昕,等. 基于深度学习理论的机械装备大数据健康监测方法[J]. 机械工程学报, 2015, 51（21）： 49-56.
LEI Yaguo, JIA Feng, ZHOU Xin, et al. A Deep Learning-based Method for Machinery Health Monitoring with Big Data[J]. Journal of Mechanical Engineering, 2015, 51（21）： 49-56.
[7]侯文擎,叶鸣,李巍华.基于改进堆叠降噪自编码的滚动轴承故障分类[J]. 机械工程学报,2018,54（7）：87-96.
HOU Wenqing, YE Ming, LI Weihua. Rolling Element Bearing Fault Classification Using Improved Stacked Denoising Autoencoders[J]. Journal of Mechanical Engineering, 2018,54（7）：87-96.
[8]关山.基于声发射信号多特征分析与融合的刀具磨损分类与预测技术[D].长春：吉林大学, 2011.
GUAN Shan. Technology of Classification and Forecast of Tools Wear Based on Multi-feature Analysis and Fusion of AE[D]. Changchun： Jilin University, 2011.
[9]王计生,喻俊馨,黄惟公. 小波包分析和支持向量机在刀具故障诊断中的应用[J]. 振动测试与诊断, 2008, 28（3）： 273-277.
WANG Jisheng, YU Junxin, HUANG Weigong. Application of Wavelet Package Analysis and Support Vector Machine to Fault Diagnosis of Cutting Tool.[J]. Journal of Vibration,Measurement & Diagnosis, 2008, 28（3）： 273-277.
[10]LI Xiaoli. A Brief Review： Acoustic Emission Methods for Tool Wear Monitoring during Turning[J]. International Journal of Machine Tools & Manufacture, 2002, 42： 157-165.
[11]谢霖铨,梁博群. 基于降噪自编码的推荐算法[J]. 计算机与现代化, 2016（2）：38-41.
XIE Linquan, LIANG Boqun. A Recommendation Algorithm Based on Denoising Autoencoders[J]. Computer and Modernization, 2016（2）：38-41.
[12]于万钧,安改换,鹿文静,等. 基于栈式降噪自动编码器的气体识别[J]. 计算机工程与设计, 2017,38（3）：814-818.
YU Wanjun, AN Gaihuan, LU Wenjing, et al. Gas Recognition Based on Stacked Denoising Autoencoders[J]. Computer Engineering and Design, 2017,38（3）：814-818.
[13]秦浩然,林基明,王俊义.基于堆叠降噪自编码的恒星/星系分类研究[J].天文学报,2016,57（3）：344-352.
QIN Haoran, LIN Jiming, WANG Junyi. Stacked Denoising Autoencoders Applied to Star/Galaxy Classification[J]. Acta Astronomica Sinica, 2016,57（3）：344-352.
[14]徐德荣,陈秀宏,田进. 稀疏自编码和Softmax回归的快速高效特征学习[J]. 传感器与微系统, 2017,36（5）：55-58.
XU Derong, CHEN Xiuhong, TIAN Jin. Fast and Efficient Feature Learning Algorithm Based on Sparse Autoencoder and Softmax Regression[J]. Transducer and Microsystem Technologies, 2017,36（5）：55-58.
[15]关山,聂鹏.在线金属切削刀具磨损状态监测研究的回顾与展望Ⅲ：模式识别方法[J]. 机床与液压, 2012, 40（3）： 148-153.
GUAN Shan, NIE Peng. The Review and Perspective of the Research of On-line and Indirect Metal Cutting Tool Condition Monitoring Ⅲ：Pattern Recognition Methods[J]. Machine Tool & Hydraulics, 2012, 40（3）： 148-153.
[16]王国锋,李启铭,秦旭达,等. 支持向量机在刀具磨损多状态监测中的应用[J]. 天津大学学报, 2011, 44（1）： 35-39.
WANG Guofeng, LI Qiming, QIN Xuda, et al. Application of Support-vector-machine in Tool Wear of Multi-stage Monitoring[J]. Journal of Tianjin University, 2011, 44（1）： 35-39.