Modeling Of Cutting Residual Stress And Distortion Prediction For Machining Thin-walled Structure

The demand for high thrust-weight ratio and high reliability puts forward higher requirements to manufacturing technology of aviation engine from three aspects of macroscopic geometry,microtopography,and surface physical features.The surface physical feature of parts becomes one of the most uncontrollable output characteristics in cutting process due to its high complexity of physical mechanism.As one of the most important indicators,cutting residual stress is extremely sensitive to process parameters,and involves elastic-plastic mechanics,thermodynamics,and interdisciplinary crystal materials at macro and micro level.Moreover,the plastic theory under large deformation with high temperature and high strain rate is not complete.All of those factors make its formation mechanism confusing.Distortion induced by cutting residual stress is a viscoelastic problem,which is a traditional problem in manufacture of thin-walled structures.This work is focused on cutting residual stress and distortion of thinwalled structure of typical aviation materials.Starting from modeling plastic constitutive relation of workpiece material,the contact behavior between cutter and workpiece is subsequently analyzed.Then,the radial basis function neural network(RBF)and the grey theory are introduced to predict and optimize cutting residual stress.Finnaly,finite element method is applied to simulate the distortion induced by cutting residual stress.The main works are as follows:(1)An inverse determination method of constitutive model of workpiece material for metal cutting has been developed.Based on the Oxley shear theory,a modified parallel-sided shear zone model is proposed.Subsequently,the strain,strain rate,and temperature on the shear band have been successfully predicted.And then equivalent stress on shear plane can be calculated by constitutive equation.The equivalent stress is simultaneously obtained from measured cutting forces.Inspired reverse thinking,the determining process of constitutive parameters is taken as an optimization problem.The constitutive model is determined by iterative computation with intelligent optimization algorithm when the difference between measured and predicted equivalent stress gets to its minimum.Case studies with two materials of high temperature alloy Inconel 718 and titanium alloy Ti4Al6 V have been conducted.The results show that the proposed constitutive models have higher prediction accuracy compared with previous models.(2)A theoretical prediction model of orthogonal cutting residual stress has been proposed.Based on the modified parallel-sided shear zone model,the cutting forces are predicted by iterative computing.The predicted cutting forces agree well with the measured value for orthogonal cutting GH4169 and TC4 materials.Then,the internal temperature field of workpiece is calculated with the predicted cutting forces based on moving heat source method and subsequently verificated with Finite element(FA)simulation.The mechanical and thermal stresses in workpiece are predicted based on the Hertz rolling and sliding contact model.With incremental plastic theory,the residual stress can be thus obtained considering material kinematic hardening by the hybrid algorithm developed by McDowell.The residual stress for orthogonal cutting TC4 material is predicted with the obtained modified J-C equation.The results show that the predicted residual stresses agree well with the measured results.(3)The RBF and grey relational analysis(GRA)have been applied on prediction and optimization of cutting residual stress.A RBF model for residual stress induced by multi-axis machining GH4169 is developed.Vefication experiments show that it has higher prediction accuracy compared with multiple linear regression method and Back-Propagation neural network(BP).The Taguchi-Grey method coupled with subjective and objective weight factor analysis is used to optimize residual stress and surface roughness for end milling titanium alloy TC18.Subsequently,a continuous multi-objective optimization algorithm is developed by integrating RBF and Fireworm Algorithm(FA)into Taguchi-Grey method.This method is then utilized to optimize cutting parameters and cutter geometric parameters for multi-axis machining GH4169 in terms of residual stress and surface roughness.(4)A prediction method of distortion induced by residual stress for machining thin-wall structure has been developed.Due to the mechanism analysis,cutting residual stress profile can be summarized into two categories.And two parametric representation models are accordingly proposed.The empirical mapping models between model coefficiets and process parameters are then built based on test data sample set.And the relation model between residual stress profile and process parameters is therefore built.The residual stress profiles for cutting TC4 and GH4169 have been successfully predicted with this method.Then subroutine SIGNI is called to model the finite element(FA)simulation of distortion of thin-wall plate and blade induced by milling operation on the ABAQUS software platform.