节点文献
航空铝合金残余应力及切削加工变形研究
Residual Stresses and Deformations of Aerospace Aluminum Alloy in Machining
【作者】 唐志涛;
【导师】 刘战强;
【作者基本信息】 山东大学 , 机械制造及其自动化, 2008, 博士
【摘要】 航空整体结构件具有结构复杂、尺寸大、材料去除率高、薄壁部位多、刚性差等特点,在加工过后,零件通常会出现弯曲、扭曲或弯扭组合等变形,加工精度难以达到设计要求。本文以航空铝合金7050-T7451为研究对象,围绕引起整体结构件加工变形的主要因素,即毛坯初始残余应力的释放与重分布,铣削加工引入的残余应力,以及铣削力、铣削热和装夹载荷的作用,借助理论分析、实验研究和数值模拟等研究手段,研究影响变形的各因素在加工的各阶段所起的作用及引起的变形形式,明确整体结构件加工变形的产生机理,从而为航空整体结构件的加工变形的控制和工艺参数优化提供理论依据。分析航空铝合金预拉伸板7050-T7451内部残余应力的产生机理,指出毛坯初始残余应力主要在热轧和固溶化淬火工艺过程中产生,采用预拉伸的方法进行消减;基于裂纹柔度法测量铝合金厚板内部残余应力的分布规律,采用有限元法计算得到测试试样的裂纹柔度函数,在分析应力计算不确定度来源的基础上,研究裂纹柔度法中插值函数及其阶数的选择对应力计算结果不确定度的影响,基于应力不确定度的最小化目标,确定9阶勒让德多项式是拟合铝合金预拉伸板内部残余应力的一种理想的插值函数及其阶数;计算得到45mm厚铝合金预拉伸板7050-T7451内部残余应力分布规律,结果表明板内残余应力呈明显的外压内拉的“M型曲线”分布,为分析与预测毛坯初始残余应力引起的加工变形问题奠定基础。采用DEFORM-2D和DEFORM-3D有限元软件,建立二维正交切削加工和三维斜角切削加工航空铝合金仿真模型。采用Johnson-Cook模型描述材料的流动应力行为,采用单元自适应网格重划技术与分离准则相结合实现切屑与工件的分离,选用无量纲Cockcroft&Latham断裂准则实现材料的断裂,摩擦模型的建立通过划分粘结区和滑动区,分别应用不同的摩擦系数来描述。设计正交切削力实验,验证有限元模型的正确性。基于仿真模型预测热-力耦合作用下工件、切屑中的非均匀应变场、应力场、温度场以及切削力,分析后刀面磨损量、刃倾角、刀刃钝圆半径等刀具几何参数对切削力和切削温度场分布的影响。基于Doelle-Hauk法测量铣削加工航空铝合金工件表面残余应力的状态,结果表明:应力主平面与试样表面基本平行,铣削加工铝合金表面残余应力近似处于二维平面应力状态;采用X射线衍射和电解抛光逐层剥除相结合的方法,测试铣削加工航空铝合金残余应力沿层深的分布规律,分析主轴转速、进给速度、后刀面磨损量对加工残余应力分布规律的影响,结果表明:铣削加工引入的残余应力沿层深的分布与后刀面磨损量密切相关,其次是主轴转速,而每齿进给量对残余应力的影响不大;建立三维双刃斜角变切屑厚度有限元模型,采用Kistler测力仪获得铣削过程中的动态切削力分布,采用红外热像仪测试得到铣削过程中切屑中的最高温度,通过有限元分析与试验研究相结合,获得不同切削条件下的工件已加工表面温度场的分布规律。基于热-力耦合理论,对铣削加工残余应力的产生机理进行解释。建立航空铝合金初始残余应力的释放与重分布引起加工变形的理论解析模型与有限元模型,基于有限元法分析材料切削去除过程中双向毛坯初始残余应力的释放与重分布引起的矩形板工件加工变形规律,结果表明:毛坯初始残余应力的释放与重分布造成了工件整体上的弯曲变形;采用组合函数近似表达铣削加工引入的残余应力,解决了有限元模型中加工残余应力的自平衡问题。研究铣削加工引入的残余应力引起的加工变形规律,得出结论:加工引入的残余拉伸或压缩应力造成了工件整体上的弯曲变形,加工残余剪切应力造成了工件整体上的扭转变形,且扭转变形量大于弯曲变形量;研究多框体零件隔框加工顺序以及双面结构零件加工工艺对加工变形的影响,基于加工变形的最小化目标,得到最优的隔框加工顺序方案及工艺路线方案。综合考虑毛坯初始残余应力、装夹效应、铣削机械载荷、铣削热载荷、加工引入的残余应力对加工变形的影响,建立多因素耦合作用下工件加工变形的预报模型;设计框类结构件的高速铣削加工试验,采用三坐标测量机测量工件的加工变形,通过有限元模拟结果与试验结果的比较,验证预报模型的正确性。多因素耦合变形预报模型的建立为进一步研究加工变形的控制技术提供依据。
【Abstract】 Monolithic components are being widely used for aerospace applications because of their combination of lower height, higher assembly quality and higher structural efficiency. However, these components are large in size, complex in structure, and are often thin-walled. In the milling process of monolithic components, more than 90% of the materials would be removed which result in their low rigidity. Therefore, the large machining deformations of the monolithic components are often observed. Such machining deformations can be attributed to the following factors: (1) the release and redistribution of the original residual stresses, (2) machining-induced residual stresses, (3) the action of dynamic cutting loads and clamping forces. Considering above influencing factors, a method combined experiment with theoretical model and computer simulation was proposed in this dissertation to study machining deformation mechanism of monolithic components, which could provide a theoretical guide for predicting and controlling the machining deformations of monolithic components.The rolling and quenching processes result in high residual stresses during the production of 7050-T7451 aluminum alloy pre-stretched plate. These residual stresses are relieved by applying a uniform plastic strain in the rolling direction. In order to measure a full through-thickness original residual stress profile of aluminum alloy thick plate, a method named as the crack compliance method was presented. The compliance functions were calculated by finite element method. An optimal expansion order was obtained based on minimizing the total stress uncertainty which was evaluated by considering the two main sources in calculation of stress uncertainty: the random errors in strain data and model error. Then the residual stresses depth profiles in pre-stretched aluminum alloy plate 7050-T7451 were determined. The results revealed that Legendre polynomials that the fit order is 9 can evaluate accurately through-thickness residual stresses of aluminum thick plate. It is also shown that the residual stresses distribution in pre-stretched aluminum alloy plate can be revealed by "M type curves", that is, the exteriors are residual compressive stress and the interiors are residual tensile stress. These works can provide a basis for analyzing and predicting the effect of the blank’s original residual stresses on machining deformations of monolithic components.Based on finite element software DEF0RM-2D and DEF0RM-3D, a two-dimensional orthogonal cutting model and a three-dimensional oblique cutting model for aerospace aluminum alloy were built, respectively. The material’s flow stress behavior was described with Johnson-Cook constitutive equation. The separation of the chips with the workpiece was realized by the combination of adaptive remeshing technique and separation criterion. The material’s failure was defined by adopting Cockcroft & Latham fracture criterion. The tool-chip friction model was the combination of a Coulomb friction model and shear (sticking) friction model. To validate the finite element model, orthogonal cutting tests were conducted. The simulation models can predict non-uniform stresses, strains distribution in the chips and the workpiece. The effects of tool geometrical parameters such as flank wear, cutting edge inclination and corner radius on cutting forces and cutting temperature fields were analyzed by three-dimensional oblique finite element model.The superficial residual stresses in milling aerospace aluminum alloy were measured by Doelle-Hauk method. The measurement results show that the principal plane is approximately parallel with the machined surface of the workpiece, which reveals that these stresses are under two-dimensional plane stress state. In the experiments, the machining-induced residual stresses in depth were measured by using X-ray diffraction technique in combination with electro-polishing technique. Particular attention was paid to the influence of cutting parameters, such as the spindle speed, feed rate and flank wear on residual stresses in milling aluminum alloy. The results revealed that the flank wear would appear to influence on machining-induced stresses most strongly, and spindle speed appear to be of minor influence on machining-induced stresses, while the feed rate could have a few less effect on machining-induced stresses. In order to correlate the residual stresses with the thermal and mechanical phenomena developed during milling, the orthogonal components of the cutting forces were measured using a Kistler 9257A type three-component piezoelectric dynamometer. The temperature fields of the machined workpiece surface were obtained with the combination of infrared thermal imaging system and finite element method. The formation of the residual stresses can be explained by thermo-mechanical coupling effects.The machining deformations caused by the release and redistribution of original residual stresses were studied by the theoretical analysis and finite element method. The research results show that the release and redistribution of original residual stresses result in bulking distortion of the plate part. To satisfy force and moment equilibrium in the finite element model, nine basis functions were chosen to represent the machining-induced stresses profile through the depth in the thin machined specimens. The machining deformations caused by machining-induced stresses were analyzed by finite element method. The simulation results show that machining-induced residual tensile and compressive stresses induce also bulking distortion of the frame component, and machining-induced residual shear stresses cause pure twisting distortion of the frame component. A finite model was built to study the effects of bulkhead processing sequences and processing routes on machining deformation of multi-frame monolithic components. An optimal processing scheme was then proposed based on minimizing the machining deformationThe machining deformations prediction model was developed considering multi-factors coupling effects including original residual stresses, clamping load, milling mechanical loads, milling thermal loads and machining-induced residual stresses. The machining deformation of a frame monolithic component was predicted by this model. To validate the prediction model, a true frame component was machined and its deformation was measured on a Coordinate Measuring Machine. The deformations by prediction model show a good agreement with the experiment results. The machining deformations prediction model can provide an effective way to study further control strategies of the machining deformations for monolithic component.This project is supported by National Natural Science Foundation of China under Grant No.50435020.