The characteristics of the acoustic emission signal during the tool wear pr
ocess in end milling are analyzed, and a signal processing scheme for abstr
acting the mean time domain averaging deviation of the signal to monitor to
ol wear is proposed. Experiments indicate that the mean deviation value is
sensitive to flank wear and its normalized value is not as dependent on mil
ling parameters as the acoustic emission root mean square signal.