Modeling Of Time-varying Cutting Temperature Field And Residual Stress Mechanism During Dry Whirlwind Milling Of Screw

Dry whirlwind milling is a typical green cutting process which is suitable for mass production of screw parts.This cutting process has a wide range of applications in the high-grade precision CNC machine tool industry and large-scale equipment manufacturing industry in high-quality screw forming processing.In dry whirlwind milling of screw,the high temperature generated in the cutting area will affect the residual stress of the workpiece.The residual stress of the workpiece will eventually affect fatigue life and corrosion resistance in the use phase.At present,the researches on cutting temperature and residual stress are mainly focused on traditional cutting technology.However,few researches have been carried out on dry whirlwind milling of screw,which has the characteristics of multi-edge intermittent forming.Particularly,there is a lack of basic theoretical research on time-varying cutting temperature field and residual stress between workpiece surface and subsurface.In this paper,a time-varying cutting temperature model is established to study the formation mechanism of residual stress.The research contents are as follows:(1)The process characteristics of multi-edge intermittent forming in dry whirlwind milling of screw are analyzed.The two-dimensional time-varying characteristics of undeformed chips in the first cutting stage and the second cutting stage are modeled based on the multi-edge intermittent geometric forming characteristics.On this basis,the characteristics of dynamic cutting force,time-varying heat source and surface topography of dry whirlwind milling of screw are analyzed.The prediction results of dynamic cutting force and surface topography are verified by cutting force experiment and surface topography measurement experiment.(2)According to the characteristics of time-varying intermittent cutting in dry whirlwind milling of screw,an analytical model of cutting temperature field is established based on the influence of time-varying heat source and dynamic cutting force.Based on the analysis of two dimensional and time-varying characteristics of undeformed chip and time-varying heat source characteristics,the time-varying heat source models of shear zone and chip friction zone are established respectively.Combined with dynamic cutting force characteristics,the time-varying temperature field models of workpiece,tool and chip in the cutting area of dry whirlwind milling of screw are established.The prediction results of the time-varying temperature field model are verified based on the cutting temperature experiment.(3)A theoretical prediction model of residual stress is developed and the optimization strategy is proposed by considering the effects of thermal stress and surface topography.Based on the established dynamic cutting force and time-varying temperature field model,the mechanical stress and thermal stress at different positions of the surface topography of workpiece are analyzed.Furtherly,the theoretical model of residual stress affected by the surface topography of the workpiece in dry whirlwind milling of screw is established.On this basis,the distributions of residual stress between the surface and subsurface of the workpiece are studied by analyzing the process of stress loading and stress release.Furthermore,the influencing factors of residual stress are analyzed,and the optimization strategy of process parameters is proposed.(4)The prediction accuracy of the mechanism modeling method is not high when it is used to predict residual stress.Therefore,an empirical prediction method is proposed to predict the residual stress of the workpiece in dry whirlwind milling of screw.The residual stress measurement experiment is carried out based on the analysis of the multi-process parameters which affect the residual stress in dry whirlwind milling of screw.The surface residual stress and average residual stress of screw workpiece are predicted by the empirical modeling method.Then the prediction performance of BP neural network,radial basis function neural network and support vector machine are compared and analyzed.