Abstract:For tool wear fault diagnosis, a CNC tools wearing fault diagnosis method was proposed based on empirical mode decomposition(EMD) for signal processing and hidden Markov model(HMM) for pattern recognition. In signal processing stage, empirical mode decomposition was done to vibration signal in the machining process, and then a number of intrinsic mode functions (IMF) were gotten, the energy values of the IMFs were calculated and the first few highenergy stage IMFs were chosen as identification parameters. In pattern recognition stage, HMMs were gotten by training the samples with basic HMM training methods, then the test samples were used to verify the accuracy of the model. After testing, the model can describe the correspondence between the machine tool status and vibration signal of machining tools, which can be applied to the monitoring and identification of tool wear.