针对精密孔锉肖帅口工过程中易出现颤振、导致精密孔表面质量下降，如何能快速、准确识别出颤振征兆发生问题，提出基于独立分量分析(ICA)的锉削振动信号信噪分离方法，以实现对镬削颤振征兆信号的快速分离。该方法据颤振信号的时频特点，利用经验模态分解(EMD)对铿削振动信号进行分解 对EMD分解所得各本征模态分量(IMF)构造出的虚拟通道进行ICA分析，分离出包含颤振发生征兆的信号。实验结果表明，利用EMD和ICA对镬削振动信号进行分解处理，可快速分离出撞削颤振征兆信号，为后续颤振识别预报及抑制环节提供基础，从而有效提高精密孔的表面加工质量。
For precision hole Xiao Shuai port workers filing process prone to chatter, resulting in decreased precision bore surface quality, and how to quickly and accurately identify the signs occur flutter problem, based on independent component analysis (ICA) of the signal to noise vibration signal cutting and filing separation, in order to achieve rapid symptom pan cut flutter signal separation. The method according to the time-frequency dither signal characteristics, the use of empirical mode decomposition (EMD) for Keng cut vibration signal decomposition levy on the proceeds of EMD modal component of each of the (IMF) constructed ICA virtual channel analysis, separation a signal containing signs flutter occurs. Experimental results show that the use of EMD and ICA to wok cut vibration signal decomposition process can be quickly isolated signs of signal collision cut flutter, flutter provide the basis for subsequent identification forecast and suppression links, thus effectively improving (2013-11-27, PDF, 1313KB, 下载17次)