EMD和SVM在刀具故障诊断中的应用

与传统方法相比,声发射传感器在刀具故障诊断方面有很大的优势。将声发射传感器应用于刀具切削过程中,提出了基于经验模态分解（EMD）和支持向量机（SVM）的刀具故障诊断方法。该方法首先对标准化的声发射信号进行经验模态分解,将分解后的有限个固有模态函数（IMF）通过一定的削减算法增强故障类型特征,把每个IMF和残余项的能量以及整个信号的削减比作为特征向量,最后将特征向量输入支持向量机进行训练和测试,判断刀具的故障类型。通过对某一刀具的故障诊断结果进行分析,验证了该方法的实用性和有效性。

Applications of EMD and SVM for tool wear fault diagnosis

Acoustic Emission（AE） sensor possesses better performance for tool wear identifying than conventional methods.In this paper,AE sensor is employed to cutting tool wear identification and a fault diagnosis approach based on empirical mode decomposition（EMD） method and support vector machines（SVM） is proposed.First,the EMD method was used to decompose the standard AE signal into a serial of intrinsic mode function（IMF） components and a residual component.Second,with a certain cutting algorithm,the IMFs with fault character were strengthened.Then,we can extract the energy of each IMF and calculate the average cutting ratio of all the IMFs and residual component,which is served as the fault characteristic vectors to be input to the support vector machine classifier.Finally,it can recognize the status of the tool wear with the SVM.The result shows that it has good performance to recognize and diagnose the tool wear,it is testified suitable to monitor the cutting tool wearness.