首页 | 本学科首页   官方微博 | 高级检索  
     检索      

应用模式识别技术判别刀具钝化
引用本文:李莉,刘又午,陈志荣,彭泽民.应用模式识别技术判别刀具钝化[J].天津大学学报(自然科学与工程技术版),1988(4).
作者姓名:李莉  刘又午  陈志荣  彭泽民
作者单位:天津大学机械制造工程系 (李莉,刘又午,陈志荣),天津大学机械制造工程系(彭泽民)
摘    要:本文采用统计模式识别的方法,从切削过程的动态切削力和振动信号中抽取特征,进行线性分类,对刀具的磨损状态进行了有效的判别。文中提出一个衡量大样本单个特征线性可分性的计算公式。在实际应用中,根据单个特征的实际分布图,选择可分性好的特征;同时,通过编辑样本选择训练集,使正确判别率得到显著提高。这种方法兼有计算量小、速度快等优点,可使用微型计算机,便于实际应用,并有可能用于在线监控。

关 键 词:模式识别  编辑样本  刀具钝化

DISCRIMINATION OF CUTTING TOOL DULLING USING PATTERN RECOGNITION TECHNIQUE
Li Li Liu Youwu Chen Zhirong Peng Zeming.DISCRIMINATION OF CUTTING TOOL DULLING USING PATTERN RECOGNITION TECHNIQUE[J].Journal of Tianjin University(Science and Technology),1988(4).
Authors:Li Li Liu Youwu Chen Zhirong Peng Zeming
Institution:Department of Mechanical Engineering
Abstract:A siatistical Pattern recognition method for cutting tool dulling is pressures. Signals ot dynamic cuiting forces and accereration during cutting motion are used for feature selection and linear classifieation. The prscedure of disorimination is effective.A calculating formula for the judgement of separabilitry of single feature wtih large sample is presented, ton.For application features with ideal separability are selected in accordance with the distribution histograms of these features. At the same time, the tats of correct recognition is increased obviously by using editing samples as training samplc.The proposed method has the advantages of reduced computer storage and (?)ased processing time. Mieroeomputers can be used, It can used for on-line tool dulling monitoring.
Keywords:pattern recognition  editing sample  cutting tool dulling
本文献已被 CNKI 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号