Abstract: The acoustic emission signal sensing of nickel-based crystal microgrinding was measured. The monitoring system of acoustic emission signal is established, the characteristic parameters are extracted from the signal and the division of frequency band is completed. The correlation degree between acoustic emission signal time domain root mean square spectrum and frequency band signal energy ratio predicted unsteady grinding characteristics is deeply discussed. The results show that the wear form of the micro-abrasive tool changes from the initial rapid wear process to the stable wear state after the grinding length increases gradually. With the increase of material removal amount, a larger diameter loss amount was formed, and the overall performance was first rising and then stable state change. There is a positive correlation between the energy ratio of middle band and the diameter loss of microabrasives. The energy ratio of high frequency band is positively correlated with the width of micro-slot edge breakage. Microgroove taper Angle usually presents continuous change characteristics, forming a larger microgroove taper angle. With the increasing of grinding depth, the diameter loss of micro-abrasive tools and the correlation coefficient of energy ratio in middle frequency band are increased, while the correlation coefficient of micro-groove edge breakage width and energy ratio in high frequency band are decreased.