Abstract:After presenting the basic theory and algorithm of Hilbert-Huang Transform (HHT), the signal is decomposed through the empirical mode decomposition (EMD) method to get the intrinsic mode function (IMF) in order to obtain the average amplitude. The IMF component which, related with tool, is chosen through the difference of screening. Meanwhile, the marginal spectrum of single intrinsic mode function is obtained and its maximum amplitude is then found. By establishing the mapping relationship with tool wear, the feature extraction is achieved. Regarding them as the input vector of Neural Network, and combined with the Hilbert spectra, the tool wear judgment is being processed. The studies shows that this approach can be a simple and reliable method for detecting the level of tool wear.