Publications

Abstract : Tool wear monitoring is an indispensable peremptorily of advanced manufacturing in order to evolve an automated unmanned production. Continuously machining with a worn or impaired tool will result in damage to the work piece. This difficulty becomes more important in subsidiary machining processes like milling which the tool has regularly passed a lot of machining processes and any destruction to work piece at these level consequences in more production losses. In this work, vibration signals in milling process are recorded and examined carefully in order to detect tool wear. The online acquiring of machined surface images has been done at intervals and those captured periodic texture of machined surface images are analysed for detection of tool wear. The vibration signals and the digital images are analysed using data mining techniques, decision tree to classify the tool wear. Further, the effectiveness of fusion of sensory data from the CCD camera (Image analysis) and an accelerometer (Vibration analysis) in tool wear prediction is checked and compared.