Tool wear monitoring with wavelet packet transform—fuzzy clustering method

Abstract In the manufacturing systems such as flexible manufacturing system (FMS), one of the most important issues is to detect tool wear under given cutting conditions as accurately as possible. This paper develops a device for detecting acoustic emission (AE) signal form rotating tool with magnetofluid and presents a method of tool wear monitoring, the method consists of wavelet packet transform preprocessor for generating features from AE signal, followed by fuzzy clustering method (FCM) for associating the preprocessor outputs with the appropriate decisions. A wavelet packet transform is used to decompose AE signal into different frequency bands in time domain, the root mean square (RMS) values extracted from the decomposed signal of each frequency band were used as feature. Analyzing the above features, the features that are directly relation to tool wear are used as final monitoring features. According to boring tool wear grades, the tool wear states were divided into ‘A’, ‘B’, ‘C’ and ‘D’ classifications, the state ‘D’ is proposed to be used as the prediction of tool replacement. FCM was proposed to classify monitoring features automatically so as to recognize tool wear status. The experimental results indicate that the monitoring features had a low sensitivity to changes of the cutting conditions and FCM has a high monitoring success rate in a wide range of cutting conditions.