摘要
基于振动信号小波包分解理论对不平稳信号特征提取的优势,提出了一种利用振动信号的能量变化来监测刀具磨损状态的方法。该方法利用db4小波基对振动信号进行4层小波包分解,并将分解后的各频带能量值作为刀具磨损状态判断的特征参数。在新刀和刀具磨损的状态下提取特征向量,并根据频段能量的变化判断刀具磨损程度。试验结果证明该方法在刀具磨损状态判断中的可行性。
Based on the advantage of the wavelet packet transform theory in the non - stationary signal feature extraction, a method is proposed to judge tool wear state by means of the energy variety of the vibration signals. This method uses db4 wavelet packet to decompose vibration signals into 4 level, and taking the decomposition of the each frequency band energy as characteristic parameters of the tool wear state. The characteristic vector is extracted under new tool condition and tool wear condition, and using this vector to judge the tool wear variety. The usability of this method is verified by the test results.
出处
《组合机床与自动化加工技术》
北大核心
2013年第6期65-67,共3页
Modular Machine Tool & Automatic Manufacturing Technique
基金
国家科技重大专项课题-高档数控机床与基础制造装备(2012ZX04005-021)
关键词
刀具磨损
小波包分解
振动信号
特征提取
tool wear
wavelet packet decomposition
vibration signals
feature extraction
