刀具声发射信号的组合去噪及特征分析  

Combined De-Noising and Characteristic Analysis of Tool's Acoustic Emission Signal

作　　者：席剑辉 许廿 XI Jian-hui;XU Nian(School of Automation,Shenyang Aerospace University,Liaoning Shenyang 110136,China)

机构地区：[1]沈阳航空航天大学自动化学院,辽宁沈阳110036

出　　处：《机械设计与制造》2018年第9期178-181,共4页Machinery Design & Manufacture

基　　金：国家自然科学基金青年基金资助项目(60804025);辽宁省教育厅科学技术研究项目(L2014069;L2013070);沈阳市科技创新团队项目(src201204)

摘　　要：针对刀具切削加工过程产生的声发射信号进行噪音滤除,以有效监测刀具磨损情况,提高工件加工质量。研究形态滤波与集合平均经验模态分解(EEMD)的有效组合方法,在时域和频域对信号进行降噪处理。首先采用加权级联形态滤波,滤除声发射信号的尖峰脉冲干扰;进而采用EEMD分解处理后的信号,计算所得本征模态分量(IMF)的相关性以去除虚假分量,达到去噪效果。仿真实例分别对模拟加噪声发射信号和实测刀具声发射信号进行处理,并提取去噪前后信号频率特征进行比较,仿真结果说明了此方法的有效性。Aimed at filtering the noise of acoustic emission signal generated when tool is cutting,which can effectively monitor the tool wear and improve the machining quality.The ensemble average empirical mode decomposition(EEMD)and the morphological filter are effectively combined to denoise the acoustic emission signal in the time and frequency domain.Firstly use the weighted cascade morphological filtering to filter the peak pulse interference of acoustic emission signal.The processed signal was presented to the EEMD to be decomposed.According to the correlation of available intrinsic mode function(IMF),the false components could be removed and the signal de-noising could be realized.Both the simulated and the measured noisy acoustic emission signal were presented as examples.The frequency characteristics are extracted and compared between the noisy signal and the de-noising signal.Simulation results show the effectiveness of the proposed method.

关 键 词：声发射 形态滤波 EEMD 去噪 

分 类 号：TH16[机械工程—机械制造及自动化] TG711[金属学及工艺—刀具与模具]