基于谐波特征和GA-SVM的刀具状态监测

摘要

Tool-wear's monitoring plays an important part in automatic machining processes. In order to on-line monitor tool wear,a combined approach of characteristic harmonics and GA-SVM( Genetic Algorithm-Support Vector Machine) in on-line monitor tool wear is used. Harmonic features of AE signals is extracted firstly by wavelet transform, which was used as an input of SVM, and then GA is used to find the optimized parameter of SVM to build the model of tool condition through training. The experiment results showed that the model can effectively monitor the tool condition.%刀具状态的监测是实现机械加工自动化重要的一环.为了有效地捕捉刀具的状态信息,提出了一种基于谐波特征和GA-SVM(遗传-支持向量机)相结合的刀具状态监测方法.该方法运用小波变换提取AE信号的谐波特征信息,作为支持向量机的输入参数,GA寻找SVM建立刀具状态模型的最优参数,通过训练建立模型.结果表明,该方法能有效监测刀具磨损状态.