Optimization Of Micro-milling Process Based On BP Neural Network

With the increasingly serious energy shortage and environmental problems,green and energy-saving manufacturing has become one of the hot spots of the manufacturing industry.As one of the important processes of material removal,micro milling has been widely used in many fields of machinery manufacturing industry.Therefore,it is necessary for green and energy-saving manufacturing to study the problem of energy consumption in the process of processing.In the process of micro-milling,the tool wear is more serious than that of traditional milling because of the small machining scale.Therefore,in this paper,the parameter optimization in the process of micro-milling radial tire dies is studied as follows:(1)The micro-milling test of radial tire mold side plate was carried out.Based on the milling test data,the influences of spindle speed,feed per tooth and cutting depth on the specific cutting energy,tool wear and surface roughness in micro-milling were studied.The results show that increasing the spindle speed is not only beneficial to reduce the specific cutting energy,improve the surface roughness,and is also beneficial to reduce the tool wear,improve the machining efficiency,prolong the tool life;However,increasing the feed per tooth and cutting depth will decrease the specific cutting energy but worsen the surface quality and aggravate the tool wear.(2)In view of the shortcomings of the traditional BP neural network,the weight and threshold were optimized and improved by using genetic algorithm based on MATLAB software,and the GABP single objective prediction model was established.Compared with the traditional BP neural network,the prediction of high precision specific cutting energy,tool wear and surface roughness was realized.The multi-objective GABP neural network was improved to realize the parallel connection of multiple single-objective GABP neural networks,and the three-objective prediction model of specific cutting energy,tool wear and surface roughness was established.The prediction accuracy of the model was higher.(3)Based on the NSGA-?genetic algorithm,the cutting parameters optimization problem of micro-milling with the minimum cutting specific energy,tool wear and surface roughness as targets was solved.The pareto solutions focus:the spindle speed range of18508.050?19211.216 rpm,feed per tooth range 0.037?0.060 mm/z,cutting depth range of 0.300?0.600 mm.The spindle speed level is high,indicating that in the pursuit of high efficiency and energy saving,higher tool life,better surface quality requirements,should try to improve the spindle speed to achieve.(4)The pareto solution set and original data were sorted by using grey relational analysis,and a set of cutting parameters sui Tab.for machining production was finally obtained.The optimal cutting parameters were as follows:spindle speed n=19143.952rpm,feed per tooth f_z=0.038mm/z,cutting depth a_p=0.501mm.At the same time,the optimal machining results were as follows:tool wear S_VB=615.164?m²,specific cutting energy SCE=382.674J/mm³,surface roughness Ra=0.521?m.This research is helpful to optimize the process parameters of micro-milling and guide the actual production.