基于声发射频谱分析的磨削消空程研究  

Eliminating Grinding Idle Process Based on Spectrum Analysis of Acoustic Emission

作　　者：龙郁 李臻峰 宋飞虎 李静[1,2] LONG Yu;LI Zhenfeng;SONG Feihu;LI Jing(School of Mechanical Engineeing,Jiangnan University,Wuxi,Jiangsu 214122,China;Jiangsu Key Laboratory of Advanced Food Manufacturing Equipment and Technology,Jiangnan University,Wuxi,Jiangsu 214122,China)

机构地区：[1]江南大学机械工程学院,江苏无锡214122 [2]江南大学江苏省食品先进制造装备技术重点实验室,江苏无锡214122

出　　处：《轻工机械》2020年第4期54-57,63,共5页Light Industry Machinery

基　　金：国家自然科学基金(21606109)。

摘　　要：为解决由单一声发射信号时域特征作为消空程指标引起的误报,课题组设计了接触速度作为变量的单因素实验,分析磨削进程开始前后信号差异。并提出已信号频域最大峰值频率作为指标之一,采用主成分分析结合费舍尔判别法来监测磨削接触的发生,进而实现磨削消空程,研究结果表明:该方法对砂轮工件接触与否的判别的准确率达到96.7%,且与传统判别法比较,能通过高频信号变化更早识别到接触的发生。该方法提高了消空程效果,且能提前识别到接触进程的发生,为实际生产设计提供了参考。In order to eliminate the false alarms caused by the time-domain characteristics of single acoustic emission signal as the index of eliminating idle process,a single factor experiment with contact speed as a variable was designed to analyze the signal difference between idle process and grinding process. A method using principal component analysis combined with Fisher discriminant was been proposed to monitoring the beginning of process. The results show that the accuracy of the method to discriminating the contact up to 96. 7%. Compared with the traditional discrimination method,the occurrence of contact can be recognized earlier through the change of high-frequency signals. This method improves the accuracy of the eliminating grinding idle process and recognizes the occurrence of the contact process in advance,and provide reference for actual production design.

关 键 词：磨削加工 声发射 消空程 主成分分析 费舍尔判别法 

分 类 号：TP277[自动化与计算机技术—检测技术与自动化装置] TH132[自动化与计算机技术—控制科学与工程]