The tool wear status identification with the traditional feature extraction method cannot fully characterize the wear characteristics of vibration signals

thus leading to the low accuracy of wear status identification. Therefore

we propose a tool wear status identification method based on the direct fast iterative filter (dFIF) decomposition method

the Slime mould algorithm and the support vector machine (SMA and SVM). Firstly

the milling tool vibration signal is decomposed with the dFIF. Secondly

the eigen-mode function generated by the decomposition method is calculated and scored using the weighted sparse kurtosis index

and the eigen-mode function with high scores is selected for noise reduction and reconstruction. Finally

a model is constructed using the SMA and SVM. The reconstructed signal features are input into the optimized model after dimensionality reduction with the principal component analysis to classify and identify tool wear status. The experimental results prove that the tool wear status identification rate of the proposed method is as high as 99.8% and that the method can identify the four wear statuses of the milling tool quickly and accurately.