The Research On The Grinding Wheel Wear And Surface Roughnes In Acoustic Emission Monitoring Of Engineering Ceramic In Grinding

As a nondestructive detection technology,the AE(acoustic emission)technology has gained wide attention and application in the field of grinding processing.With the development trend of intelligent manufacturing,the AE technology will occupy a more important position in the field of grinding,or will become an important part of intelligent grinding system.Because of the high complexity of the grinding process,the AE signals generated from the grinding process are also very complex and contain numerous grinding information.Therefore,to explore the correspondence between the AE signals from grinding and grinding conditions,it is of great significance to realize the on-line application of the AE technology.This paper is based on common acoustic emission grinding experiments and the main research contents are as follows:This paper first introduces the research status of the AE technology in the field of grinding,as well as the mechanism of the AE technology and the processing method of the AE signals.The components of acoustic emission signal acquisition system and the functions of each part are also introduced.Based on this,an AE experiment monitoring system is constructed and three AE experiments were studied.Based on the AE experiments of broken pencil lead,the time-frequency domain characteristics of broken pencil AE signals are analyzed under analysis of discrete wavelet transform(DWT),empirical mode decomposition(EMD)and short-time Fourier transform(STFT).The results show that the three methods can separate the friction noise signal from the signal.Based on the AE experiment of grinding partially stabilized zirconia(PSZ)under different wear of grinding wheel,the AE signals of different grinding wheels were analyzed by EMD,and it is found that the intrinsic modal functions imf1?imf6 obtained by EMD have a strong correlation with the original signal.Its main frequency is close to the main frequency of the original signal,and the energy accounted for more than 85%,which can represent the original acoustic emission signal.The Root-Mean-Square(IMFrms),variance(IMFvar)and energy coefficient(IMFpe)are extracted from imf1?imf6.The correlation between these characteristic parameters and the degree of wear of the grinding wheel was found.Finally,the least square support vector machine(LS-SVM)was used to identify the wear state of the grinding wheel,and a good recognition effect was obtained.Combining the research situation of the AE technology in surface roughness of grinding,6 sets of AE signal parameters selected from the AE signals are used as characteristics vector to predict the surface roughness after PSZ grinding.Using BP neural network and BP neural network optimized by genetic algorithm(GA-BP)to predict the surface roughness,it was found that the prediction result of GA-BP neural network is more accurate.The AE grinding experiment studied in this paper is of great significance to the application of the AE technology in grinding.