2025年01月24日 星期五 首页   |    期刊介绍   |    编 委 会   |    投稿指南   |    期刊订阅   |    统合信息   |    联系我们
计量学报  2021, Vol. 42 Issue (8): 1034-1040    DOI: 10.3969/j.issn.1000-1158.2021.08.09
  力学计量 本期目录 | 过刊浏览 | 高级检索 |
基于卷积门控循环神经网络的刀具磨损状态监测
吴凤和1,2,钟浩1,章钦1,郭保苏1,2,孙迎兵1,2
1. 燕山大学机械工程学院,河北 秦皇岛 066004
2. 河北省重型智能制造装备技术创新中心,河北 秦皇岛 066004
Tool Wear Condition Monitoring Based on Convolution-gated Recurrent Neural Network
WU Feng-he1,2,ZHONG Hao1,ZHANG Qin1,GUO Bao-su1,2,SUN Ying-bin1,2
1. Mechanical College,Yanshan University,Qinhuangdao,Hebei 066004,China
2. Heavy Intelligent Manufacturing Equipment Technology Innovation Center of Hebei Province,Qinhuangdao, Hebei 066004, China
全文: PDF (2017 KB)   HTML (1 KB) 
输出: BibTeX | EndNote (RIS)      
摘要 针对刀具磨损状态在线监测需求,提出一种基于卷积门控循环神经网络的刀具磨损状态在线监测方法。综合卷积神经网络和门控循环神经网络的优点,构建了卷积门控循环神经网络;以切削力为输入信号,通过小波变换滤除噪声;利用卷积神经网络提取表征刀具磨损状态关键信息的高维特征;通过门控循环神经单元使模型在时间尺度上的累积效应得到充分表达,体现磨损的时序特性。实验表明,在有限的刀具磨损数据样本条件下,通过卷积门控循环神经网络进行刀具磨损状态监测具有较好的效果,其准确率达到97%。
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
吴凤和
钟浩
章钦
郭保苏
孙迎兵
关键词 计量学刀具磨损切削力卷积门控神经网络在线监测    
Abstract:Aiming at the demand of online monitoring tool wear status, a method for online monitoring tool wear status based on convolution gated recurrent neural network was proposed. Based on the advantages of convolutional neural network and gated recurrent unit neural network, a convolutional gated recurrent neural network was constructed; the cutting force was used as the input signal and the noise was removed by wavelet noise reduction; the convolutional neural network was used to extract the high-dimensional features that represented the key information of the tool wear status; the cumulative effect of the model on the time scale was fully expressed through the gated recurrent neural unit, which reflected the wear time-series characteristics. Experiments showed that under the condition of limited tool wear data sample, the tool wear state monitoring through convolution-gated recurrent neural network had a good effect, and the accuracy rate was 97%.
Key wordsmetrology    tool wear    cutting force    convolution gated recurrent neural network    online monitoring
收稿日期: 2019-12-16      发布日期: 2021-08-20
PACS:  TB931  
基金资助:国家重点研发计划(2016YFC0802900);国家自然科学基金青年科学基金(51605422);河北省自然科学基金(E2017203372,E2017203156);2018年河北省专业学位教学案例(库)建设项目(KCJSZ2018023);河北省教育厅在读研究生创新能力培养资助项目(CXZZSS2019038)
通讯作者: 孙迎兵(1989-),男,河北衡水人,燕山大学实验师,硕士,主要研究方向为数字化设计制造。Email:sunyingbing@ysu.edu.cn     E-mail: 332878382@qq.com
作者简介: 吴凤和(1968-),男,内蒙古扎兰屯人,燕山大学教授,博士生导师,主要从事智能制造方面的研究。Email:risingwu@ysu.edu.cn
引用本文:   
吴凤和,钟浩,章钦,郭保苏,孙迎兵. 基于卷积门控循环神经网络的刀具磨损状态监测[J]. 计量学报, 2021, 42(8): 1034-1040.
WU Feng-he,ZHONG Hao,ZHANG Qin,GUO Bao-su,SUN Ying-bin. Tool Wear Condition Monitoring Based on Convolution-gated Recurrent Neural Network. Acta Metrologica Sinica, 2021, 42(8): 1034-1040.
链接本文:  
http://jlxb.china-csm.org:81/Jwk_jlxb/CN/10.3969/j.issn.1000-1158.2021.08.09     或     http://jlxb.china-csm.org:81/Jwk_jlxb/CN/Y2021/V42/I8/1034
京ICP备:14006989号-1
版权所有 © 《计量学报》编辑部
地址:北三环东路18号(北京1413信箱)  邮编:100029 电话:(010)64271480
本系统由北京玛格泰克科技发展有限公司设计开发  技术支持:support@magtech.com.cn