Feature Extraction and Analysis Method of Spindle Vibration Signal for Tool Wear Monitoring

Tool wear monitoring plays a positive role in improving the machining quality, efficiency and reducing the cost of the cutting process. However, due to the complexity of the cutting process and environmental interference, the original signal collected by the sensor often has a large amount of data, mixed with a lot of noise and interference information unrelated to the tool state, which cannot directly and effectively reflect the change of the tool wear state. Therefore, it is necessary to de-noise the original signal and extract features. In this paper, the change of vibration signal of tool wear in the boring process is studied. Based on wavelet packet de-noising, the features of the time domain, frequency domain, and time-frequency domain are extracted, and the corresponding rules are explored in combination with the tool wear process. It is pointed out that wavelet packet de-noising can effectively remove the high-frequency noise of vibration signal, and the correlation between different features of signal and tool wear is different. It provides a method for feature selection and analysis of tool wear monitoring.