摘要
为了实现金刚石砂轮磨削加工金刚石滚轮过程的自动化,需要对磨削接触状态进行准确识别。由于磨削过程中材料去除率较小导致声发射信号幅值变化不显著,仅用有效值对磨削接触状态识别的准确性受噪声影响很大。针对此问题,通过模态分解和相关性分析相结合的方法对采集的声发射信号进行处理,再计算各分量的有效值和方差值完成特征提取,最后利用支持向量机对磨削接触状态进行识别。实际应用发现:该方法对滚轮的磨削接触状识别准确率达到了98.3%,准确实现了对磨削接触状态的识别。
For the automation of diamond roller grinding with diamond wheel,it is necessary to accurately identify the grinding contact state.Due to the low material removal rate during the grinding process,the amplitude of the acoustic emission signal does not change significantly and the accuracy of grinding contact state recognition by using only effective value is greatly affected by noise.To solve this problem,the acoustic emission signals were processed by combining modal decomposition and correlation analysis,and then the effective values and variance values of each component were calculated to complete feature extraction.Finally,support vector machine was used to identify the grinding contact state.The actual application shows that the recognition accuracy of grinding contact condition of roller is 98.3%,and the recognition accuracy of grinding contact condition is realized.
作者
赵华东
刘勇
朱振伟
张瑞
ZHAO Hua-dong;LIU Yong;ZHU Zhen-wei;ZHANG Rui(College of Mechanical and Power Engineering,Zhengzhou University,He’nan Zhengzhou 450001,China)
出处
《机械设计与制造》
北大核心
2024年第2期174-178,共5页
Machinery Design & Manufacture
基金
郑州市协同创新重大专项(18XTZX12006)。
关键词
金刚石滚轮
金刚石砂轮
声发射
模态分解
特征提取
磨削接触状态识别
Diamond Roller
Diamond Grinding Wheel
Acoustic Emission
Modal Decomposition
Feature Extrac-tion
Grinding Contact State Recognition
