FEATURE EXTRACTION AND ASSESSMENT USING WAVELET PACKETS FOR MONITORING OF MACHINING PROCESSES

Abstract Monitoring of machining processes is a classic and yet unsolved problem in manufacturing engineering. This paper introduces a new method of feature extraction and feature assessment using a wavelet packet transform for monitoring of machining processes. First, the principles of wavelet transforms are briefly described by comparing to the Fourier transform, followed by a discussion on the properties of wavelet transforms and a special wavelet transform algorithm: the wavelet packet transform. Next, based on the wavelet packet transform and signal reconstruction, a new feature extraction method is introduced. To assess the effectiveness of the selected features in both time and frequency domains, four criteria are proposed. Accordingly an automatic feature extraction procedure is developed. The proposed method is tested using two practical examples: chatter monitoring in turning and tool wear monitoring in drilling. Experimental results show that the proposed method is very effective.