Research On Prediction And Optimization Method Of Deformation Induced By Residual Stresses In Milling Of Thin-walled Parts

Thin-walled metal parts which have the characteristics of weak stiffness,complex structure and high precision requirements are widely used in aerospace industry.During the milling process,the large material removal rate and the strong thermodynamic coupling of the contact interface make severe deformation due to residual stresses.At present,there is still a lack of effective prediction and control methods for the deformation induced by residual stresses.Aiming at the milling process of thin-walled parts,this paper studied relative key issues including the residual stresses induced by milling,the prediction methods for deformation due to residual stress as well as its apllications in the cutting process optimization.1.A prediction method of the residual stress distribution induced by milling is proposed based on cutting conditions.Firstly,the formation mechanism of milling residual stress is analyzed based on the cutting process,the cutting conditions which affects the formulation of residual stresses are defined and the mapping model between the residual stresses and cutting conditions is estiblished.Then,the residual stress tests are carried out.The X-ray diffraction method is used to measure the distribution of the residual stress under different ball end milling conditions.The influence law and mechanism of the residual stress distribution on the workpiece are analyzed.Finally,an empirical prediction model of the residual stress induced by ball end milling is established based on the experimental data,which can realize the fast and accurate prediction of the residual stress in the milled workpiece.2.A prediction method of deformation is proposed based on residual stress field.For the deformation caused by initial residual stress,the deformation of the plate under two-dimensional stress filed is deduced by the equal force moment method on the basis of the initial residual stress field,and the finite element prediction model of the deformation due to initial residual stresses is established in Abaqus based on the technology of the life and death unit.The analytical prediction results are in good agreement with the finite element prediction results.For the deformation caused by machining residual stress,a prediction method is put forward based on the mapping of working conditions and the application of machining residual stress field.The machined surface of workpiece is divided into different regions according to different working conditions,and the machining residual stress fields are applied to the thin surface shells according to the region partition.On this basis,standard parts are designed and machined,the relationship between the residual stress field and the the deformation is obtained by simulation,a correction method is further proposed based on standard parts machining,which achieve a high precision prediction of milling residual stress deformation in thin-walled parts.3.A prediction method of deformation is proposed based on clamping force monitoring.First,a mathematical model of the prediction of residual stress deformation based on clamping force monitoring is established.Then,the solving method for the model is proposed on the basis of the theory of the indeterminate structure and the equivalent solution of the deformation induced by residual stresses is obtained by using the deformation coordination equation.After that,the concrete steps to predict the deformation induced by residual stresses of thin-walled parts are put forward.Then,an experimental platform for predicting the deformation based on clamping force monitoring is established.The effectiveness of the prediction method is verified.Furthermore,a finite element model of the monitoring process of clamping forces is established,in which the spring element is used to replace the contact of the sensing point.On this basis,a fast prediction method of the deformation induced by residual stresses is proposed based on BP neural network.A blade is taken as an example for verification and good predictive effect is achieved.4.Three optimization methods are proposed with the purpose of controlling the deformation caused by machining induced residual stresses.As cutting parameters have a significant influence on the machining induced residual stresses,an optimization method of cutting parameters is proposed based on GRA(gray relational analysis)-SVM(support vector machine)-GA(genetic algorithm)to maximize the machining efficiency under the restraint of residual stress deformation.Then,a symmetrical cutting process optimization strategy is put forward to reduce the effect of machining induced residual stress equivalent torque through symmetrical balance.Finally,an optimization strategy of cutting direction is proposed based on the phenomenon of different machining stress distributions in different cutting directions.5.According to the theory and methodology proposed in this paper,a prediction systerm for the deformation induced by residual stresses is developed,and an aeroengine blade is used for application verification.Application results show that the prediction method can predict the deformation induced by residual stresses of blades with high precision.