According to acoustic emission signal of cutting tool wear states, this paper presents a method of cutting tool condition identifying based on empirical mode decomposition (EMD) and Support Vector Machine (SVM). AE signal was decomposed into a series of intrinsic mode functions (Intrinsic mode function, IMF) by EMD, extract the energy of IMF as feature vector, SVM-based tool wear identifying model was constructed by learning correlation between extracted features and actual tool wear state. In the experiment, the tool wear state was divided into: normal cutting, medium wear and severe wear. This paper compared the results of wavelet packet decomposition (WPD) method shows that EMD method was more accurate than wavelet packet decomposition to extract features of tool wear. Experimental results by cutting GH536 and GH4169 show that cutting a variety of materials tool in tool wear identification, the method based on EMD and SVM can be used.
Second International Conference on Digital Manufacturing and Automation, ICDMA 2011, Zhangjiajie, Hunan, China, August 5-7, 2011; 01/2011