基于CART决策树算法的磨削砂轮钝化状态的识别
详细信息 查看官网全文
- 英文论文题名：Recognition of the dull state of grinding wheel based on CART decision tree algorithm
- 论文作者：钟利民 ; 杨京 ; 程建春 ; 刘翔雄
- 英文论文作者：ZHONG Li-ming ; YANG Jing ; CHENG Jian-chun ; LIU Xiang-xiong ; Institute of Acoustic ; Nanjing University ; Kunshan Hiecise Machine Co. ; Ltd
- 年：2016
- 作者机构：南京大学声学研究所;昆山华辰机械制造有限公司;
- 论文关键词：磨削砂轮钝化状态 ; 小波分解 ; CART决策树
- 英文论文关键词：the state of grinding wheel ; wavelet decomposition ; CART decision tree
- 会议召开时间：2016-10-28
- 会议录名称：2016年全国声学学术会议论文集
- 语种：中文
- 分类号：TG743;TG580.6
- 学会代码：OGSM
- 会议名称：2016年全国声学学术会议
- 会议地点：中国湖北武汉
- 主办单位：中国声学学会
- 学会名称：中国声学学会
- 页数：4
- 文件大小：362k
- 原文格式：D
- 会议级别：全国
摘要
在机械加工过程中磨削砂轮的钝化状态对磨削效率和加工质量有着重要的影响,如何能够在不停止磨削进程的情况下自动识别砂轮的钝化状态,以判断是否需要对砂轮进行修整,将对提高加工效率和加工质量具有重要意义。本文提出了一种利用决策回归树(CART)算法来识别砂轮钝化状态的方法。对磨削过程中产生的声发射信号使用小波分析进行多尺度分解,提取各层小波能量将其作为砂轮钝化状态所对应的多个特征,然后通过CART算法构建出一个多分类决策树,对砂轮所处的各种钝化状态实现了精确分类。
In the process of machinery, the dull state of grinding wheels is importance to the grinding efficiency and quality. The acoustic emission generated in the grinding process by wheels has a close relationship with its own material of wheels and the grinding state. Through analysis of acoustic emission, the dull state of the wheels can be clear in the course of grinding. In this article, a CART algorithm is proposed to recognize the dull state of the grinding wheels. The acoustic emission generated in the grinding process is multi-scale decomposed with the Wavelet Analysis. The various dull states of the wheels are represented by the distribution of wavelet energy. A decision tree classifier for multi-classification is built by CART algorithm. Thus, the various dull states of the wheels get classified.
In the process of machinery, the dull state of grinding wheels is importance to the grinding efficiency and quality. The acoustic emission generated in the grinding process by wheels has a close relationship with its own material of wheels and the grinding state. Through analysis of acoustic emission, the dull state of the wheels can be clear in the course of grinding. In this article, a CART algorithm is proposed to recognize the dull state of the grinding wheels. The acoustic emission generated in the grinding process is multi-scale decomposed with the Wavelet Analysis. The various dull states of the wheels are represented by the distribution of wavelet energy. A decision tree classifier for multi-classification is built by CART algorithm. Thus, the various dull states of the wheels get classified.
引文
[1]Williams R V.Acoustic emission[M].Bristol:A.Hilger,c1980:25-27.
[2]J.-S.Kwak,M.-K.Ha,Detection of dressing time using the grinding force signal based on the discrete wavelet decomposition,International Journal of Advanced Manufacturing Technology 23(2004)87-9.
[3]Xiaozhi Chen,Beizhi Li,Acoustic emission method for tool condition monitoring based on wavelet analysis,Int J Adv Manuf Technol,2007,33:968-976.
[4]Breiman L,Friedman,J,Stone C.Classification and Regression Trees.Wadsworth,1984.
[5]Quinlan JR.Induction of decision trees.Machine Learning,1986,1(1):81-106.
[6]Xiaoli Li,A brief review:acoustic emission method for tool wear monitoring during turning,International Journal of Machine Tools&Manufacture 42(2002)157–165.
[7]徐水竹,杨京,程建春等。基于小波变换的磨削噪声在线监测方法研究,声学技术,2011;30:275-279.
[8]许黎明,许开州,柴运东。Identification of Grinding Wheel Wear Signature by a Wavelet Packet Decomposition Method[J].Journal of Shanghai Jiaotong University,2010,15(3):323-328.
[9]徐晓丽,舒赜,钟秉林。模糊决策树在磨削状态识别中的应用[J].微计算机信息(测控自动化),2008,24:7-1.
[10]Polat K,Gunes S.A novel hybrid intelligent method based on C4.5decision tree classifier and one-against-all approach for multi-class classification problems.Expert Systems with Applications,2009,36:1587-1592.
[11]Sanjay Agarwal,P.Venkateswara Rao.Grinding characteristics,material removal and damage formation mechanisms in high removal rate grinding of silicon carbide.International Journal of Machine Tools&Manufacture,2010,50:1077-1087.
[2]J.-S.Kwak,M.-K.Ha,Detection of dressing time using the grinding force signal based on the discrete wavelet decomposition,International Journal of Advanced Manufacturing Technology 23(2004)87-9.
[3]Xiaozhi Chen,Beizhi Li,Acoustic emission method for tool condition monitoring based on wavelet analysis,Int J Adv Manuf Technol,2007,33:968-976.
[4]Breiman L,Friedman,J,Stone C.Classification and Regression Trees.Wadsworth,1984.
[5]Quinlan JR.Induction of decision trees.Machine Learning,1986,1(1):81-106.
[6]Xiaoli Li,A brief review:acoustic emission method for tool wear monitoring during turning,International Journal of Machine Tools&Manufacture 42(2002)157–165.
[7]徐水竹,杨京,程建春等。基于小波变换的磨削噪声在线监测方法研究,声学技术,2011;30:275-279.
[8]许黎明,许开州,柴运东。Identification of Grinding Wheel Wear Signature by a Wavelet Packet Decomposition Method[J].Journal of Shanghai Jiaotong University,2010,15(3):323-328.
[9]徐晓丽,舒赜,钟秉林。模糊决策树在磨削状态识别中的应用[J].微计算机信息(测控自动化),2008,24:7-1.
[10]Polat K,Gunes S.A novel hybrid intelligent method based on C4.5decision tree classifier and one-against-all approach for multi-class classification problems.Expert Systems with Applications,2009,36:1587-1592.
[11]Sanjay Agarwal,P.Venkateswara Rao.Grinding characteristics,material removal and damage formation mechanisms in high removal rate grinding of silicon carbide.International Journal of Machine Tools&Manufacture,2010,50:1077-1087.