Research On Adaptive Control Method Of Machining Parameters For Surface Quality

Intelligent manufacturing is the main direction of building a strong manufacturing country in China.CNC machine tools,as the "mother of industry",the intelligent level of which plays a decisive role in the promotion of intelligent manufacturing.In the traditional NC machining mode,the machining parameters are fixed,which is difficult to adapt to the complex dynamic machining process,so it's not intelligent enough.Especially for aerospace parts,because of the thin-walled structure,the traditional machining mode is easier to produce vibration and deformation,which greatly affects the machining quality and efficiency.Adaptive machining parameter control can effectively solve the above problems,which is an important embodiment of machine tool intelligence.But at present,the research of adaptive machining parameter control is more focused on machining efficiency and less on machining quality.In this paper,to ensure the surface roughness of milling process as the primary goal,the surface quality oriented adaptive machining parameter control method is studied.Firstly,aiming at the serious influence of milling chatter on surface roughness,the on-line detection and control method of milling chatter is studied.Based on the wavelet packet analysis of milling vibration signal,an on-line detection method of milling chatter is proposed,which takes the energy entropy of wavelet packet as the characteristic quantity,and can detect the milling state which just appears the chatter trend in time.Based on the analysis of milling stability conditions,the method of on-line suppression of milling chatter based on the speed of stable region is studied,and the corresponding control strategy of spindle speed in stable region is formulated.In order to control the feed rate without chatter,the prediction model of milling surface roughness and the feed rate control method based on the model are studied.After analyzing the problems in the current prediction modeling,this paper puts forward a prediction modeling method of surface roughness based on SVM classification algorithm and input of machining parameters and RMS value of vibration signal.Based on the general incremental learning method of SVM,the incremental learning method for surface roughness prediction model is improved.In order to solve the problem that the output of classification model can not be used as the basis of feed rate control directly,the concept of classification confidence is introduced,and the fuzzy controller of feed rate is designed with the SVM classification confidence as the constraint objective,which realizes the intelligent control of feed rate.Finally,based on the SINUMERIK 840 D,the adaptive machining parameter control system is developed,the key technology is studied,and the corresponding comparative experiment is designed.The results show that the method proposed in this paper can effectively suppress chatter when chatter occurs,and can achieve the balance between surface roughness and efficiency without chatter.