The basic principle of acoustic emission technology

and its applications in monitoring of grinding processes are stated.Acoustic emission（AE） signal belongs to different band when the processing conditions or parameters are different

so a self-learning method for filter band selection is proposed.The design of the self-learning monitoring and control system is given simultaneously.The monitoring and control system filters the AE signal

monitors the grinding state by root-mean-square（RMS） voltage characteristic

and feedbacks to the CNC system.The system can lessen the time of gap process

and prevent collision between grinding wheel and workpiece.Experimental results show that the system can effectively optimize the grinding process.