基于功率谱特征分析的立铣刀磨损实时监测方法

Research on Real-time Monitoring Method of Milling Cutter′s Wear Based on Power Spectrum Feature Analysis

为了从拾取的振动信号中获取刀具的磨损状态,比较了ChebyshevⅠ型滤波器和Butterworth滤波器的振幅特性,对所获信号进行降噪滤波,并对滤波后的信号进行频谱分析。经分析发现,刀具在不同磨损阶段其基频信息和倍频特征有较大变化,利用自相关函数对其进行功率谱分析后发现,刀具急剧磨损时功率谱幅值特征较初期磨损和中期磨损阶段变化显著,可以作为刀具磨损程度的特征值指标,该特征值可用于在线监测刀具的磨损故障。

In order to obtain the wear state of the tool from the picked vibration signals,the amplitude characteristics of the Chebyshev type I filter and the Butterworth filter are compared in this paper. Noise reduction filtering is performed on the acquired signals and spectrum analysis is performed on the filtered signals.The analysis shows that the tool′s fundamental frequency information and frequency multiplication features have great changes in different wear stages.After using the autocorrelation function to analyze the power spectrum,it is found that the characteristics of the power spectrum amplitude change significantly compared with the initial wear and medium wear stages when the tool wears rapidly,and can be used as a characteristic value index of the cutting-tool wear.This characteristic value can be applied to the online monitoring of wear failures of the tool.