基于体视学原理的钛合金显微组织定量分析

钛合金显微组织的多样性给定量金相学研究带来了很大困难.在分析钛合金金相学特点及热变形中显微组织演变机制的基础上,应用体视学和定量金相学的方法,结合图形的数学运算,初步建立了一套针对钛合金不同典型显微组织特征参数的定量测量分析模型与方法,为定量化研究钛合金显微组织的演变以及组织与性能之间的关系进行了探索性工作.该定量分析模型与方法具有较强的可行性,实现了包括相的体积分数、β晶粒大小、魏氏组织中α条的厚度、丛域尺寸及网篮等多种类型组织中α相的尺寸、形态、位向分布等显微特征的参数化定量描述与分析.同时,提出了对网篮等多种类型组织中α相的计算机图像处理技术——手工分离α相的原则,有效地实现了α相的图像"解体".并给出了上述显微组织特征定量分析的实例.     

基于BP人工神经网络的TC17钛合金显微组织-力学性能关系预测

钛合金的性能对其组织状态十分敏感,与组织的多种显微特征呈现非线性的交互关系。本研究在定量分析钛合金显微组织的基础上,采用BP人工神经网络方法建立了TC17钛合金组织与力学性能的关系模型。该模型输入的显微组织特征参数包括：α相体积分数、α相厚度和不同形态α相的体积分数,输出的力学性能包括抗拉强度、屈服强度、延伸率和断面收缩率。结果表明,该模型具有很好的预测精度和泛化能力。应用贝叶斯正则化和动量梯度下降学习法较好地解决了传统BP人工神经网络训练高精度和预测低精度的过拟合现象。此模型的建立对构建TC17合金利用组织预报力学性能的专家知识库具有重要作用,而且对钛合金专家系统的整体开发具有重要指导意义。     

基于BP人工神经网络的TC17钛合金显微组织-力学性能关系预测

钛合金的性能对其组织状态十分敏感，与组织的多种显微特征呈现非线性的交互关系。本研究在定量分析钛合金显微组织的基础上，采用BP人工神经网络方法建立了TC17钛合金组织与力学性能的关系模型。该模型输入的显微组织特征参数包括：α相体积分数、α相厚度和不同形态α相的体积分数，输出的力学性能包括抗拉强度、屈服强度、延伸率和断面收缩率。结果表明，该模型具有很好的预测精度和泛化能力。应用贝叶斯正则化和动量梯度下降学习法较好地解决了传统BP人工神经网络训练高精度和预测低精度的过拟合现象。此模型的建立对构建TC17合金利用组织预报力学性能的专家知识库具有重要作用，而且对钛合金专家系统的整体开发具有重要指导意义。     

Ti-6Al-4V钛合金网篮组织与拉伸性能 定量关系研究

建立钛合金显微组织-拉伸性能定量关系对于组织性能的控制及优化具有重要意义。本文基于定量金相学和体视学原理,利用多元非线性回归方法建立了Ti-6Al-4V钛合金网篮组织-拉伸性能定量关系模型。模型自变量包括片状α相厚度、片状α相长宽比,因变量为合金的室温拉伸性能。经验证表明,所建立的多元回归模型对合金的拉伸性能具有较高的预测精度,说明模型的构建方法是可行的。     

Ti-6Al-4V合金网篮组织与拉伸性能定量关系研究

建立钛合金显微组织-拉伸性能定量关系对于组织-性能的控制及优化具有重要意义。基于定量金相学和体视学原理,利用多元非线性回归方法建立了Ti-6Al-4V合金网篮组织-拉伸性能定量关系模型。模型自变量包括片状α相厚度、片状α相长宽比,因变量为合金的室温拉伸性能。经验证表明,所建立的多元回归模型对合金的拉伸性能具有较高的预测精度,预测值与实验值相吻合。     

α+β钛合金微观组织对强韧性的影响概述

α+β钛合金的微观组织对强度和断裂韧性有重要影响。本文概述了Ti-6Al-4V、Ti-62222s、Ti-6Al-2Sn-4Zr-6Mo等钛合金的拉伸性能、断裂韧性与组织特征参数的定性和定量关系。同时指出组织参数定量化、神经网络和有限元技术的综合利用可以有效预测组织和性能的关系。     

钛合金组织特征参数与拉伸性能预测

研究了TA15钛合金β区热变形时,变形温度、变形量、变形速率、冷却方式等工艺参量对片状组织特征参数β晶粒尺寸D、α集束尺寸d、α片厚b的影响,以及片状组织特征参数与力学性能的关系.利用多元回归分析得到了强度、塑性与3个特征参数D、d、b的定量关系式,通过这种定量关系可以预测力学性能.     

定量分析热处理对Ti_2AlNb合金显微组织及硬度的影响

对B2相区等温锻造后的Ti-22Al-25Nb合金在热处理中的显微组织特征参数进行定量分析。用图像分析软件Image-Pro Plus测量了各个相的体积分数、初生粗板条O相以及次生细板条O相的宽度及长度,采用多元线性回归的方法建立了显微组织特征参数与显微硬度的定量关系模型。结果表明:合金的显微组织主要取决于热处理制度。粗板条状O相通过固溶处理得到,较细的次生针状O相通过时效处理调节。初生粗板条O相的宽度和体积分数通过固溶处理控制;而二次析出的针状O相通过时效处理来调节。实验结果与统计分析表明,显微组织特征参数与显微硬度之间满足线性关系。     

基于数字图像处理的TC4钛合金金相组织定量分析

通过对TC4钛合金显微组织的金相图片的研究,提出一种对金相定量分析的数字图像处理方法。以ImageJ软件为工具,采用数字图像处理中的图像增强、阈值化、图像分割等数字化处理方法,计算出钛合金组织在不同热处理状态下α相所占比例,以及金相分析中其他的特征参数。研究结果表明:该数字图像处理方法可实现对TC4组织金相图片的处理,在TC4棒材锻件中α相占49.38%,长短轴比为1.458。     

基于多轴载荷相位差的神经网络预测钛合金疲劳寿命EI北大核心CSCD

钛合金是航空、航天等领域重要的结构材料,面临着多轴疲劳寿命预测困难与繁琐等问题。为此,本文考虑了对称正弦波加载条件下的相位差与加载路径非比例度的关系,构建一种以相位差、正应变幅值和切应变幅值作为输入变量,以疲劳寿命作为输出变量的多轴疲劳寿命预测的神经网络模型;分别以18组纯钛、14组BT9钛合金和33组TC4钛合金多轴疲劳试验数据进行训练,用另外的2组纯钛、2组BT9钛合金和5组TC4钛合金数据进行测试;最后与等效应变模型、最大切应变模型、临界面模型进行比较。结果表明:与基于实验数据的唯象预测方法相比,基于多轴载荷相位差的神经网络方法在预测钛合金疲劳寿命方面更高效、更准确、更具普适性。     

Modelling of 7050 aluminium alloy friction stir welding

Abstract

A thermal model combined with a microstructural and yield strength model has been developed to give a prediction of precipitate evolution and strength in the as welded and post-weld heat treated condition for friction stir welding of 7xxx aerospace aluminium alloys. This fully coupled model is applied to an overaged high strength 7050 aluminium alloy friction stir welded using a range of welding rotation and translation speeds. The evolution of the microstructure has been predicted as a function of the process parameters. The resulting microstructural evolution is shown to be a complex function of both peak temperature observed during the weld cycle and heating/cooling rates. Yield strength has been calculated from the microstructural predictions and a comparison between predicted yield strength and measured hardness has been used to test the modelling approach. Reasonably good agreement between model and experiment is found over the wide range of process parameters investigated.     

基于ANN的TA15钛合金TIG焊接头拉伸性能模型建立

近α型钛合金TAI5具有较好的综合力学性能，在飞机结构中有广阔的应用前景。针对TAI5钛合金进行了系统的TIG焊试验，并进行了焊接接头拉伸性能测试。在此基础上，基于人工神经网络（ANN）原理对影响TAI5钛合金TIG焊接头拉伸性能的主要参数进行了选取，确定ANN的输入和输出参数，建立了TAI5钛合金TIG焊接头拉伸性能ANN模型并对模型进行了优化及误差分析。结果表明，最终建立的拉伸性能模型能很好地适用于TAI5钛合金TIG焊接头拉伸性能的模拟，优于传统回归方法。     

The evolution of microstructure during the processing of gamma Ti−Al alloys

This paper describes titanium research at the Queen's University of Belfast, particularly focusing on the successful combination of experimental and computer modeling techniques. Experimental work combined x-ray diffraction, optical microscopy, and scanning-electron microscopy performed on a Ti-46Al-1.9Cr-3Nb alloy before and after various heat treatments. A phase-field model was developed to simulate the formation and evolution of lamellar microstructure in TiAl alloys. In addition, a model was developed to predict the correlation between alloy composition and microstructure and its tensile properties in gamma-based titanium aluminide alloys through the use of the artificial neural network.     

Prediction of Tensile Property of Hydrogenated Ti600 Titanium Alloy Using Artificial Neural Network

An artificial neural network (ANN) model has been developed to analyze and predict the correlation between tensile property and hydrogenation temperature and hydrogen content of hydrogenated Ti600 titanium alloy. The input parameters of the neural network model are hydrogenation temperature and hydrogen content. The output is ultimate tensile strength. The accuracy of ANN model was tested by the testing data samples. The prediction capability of ANN model was compared with the multiple linear regression approach and response surface method. The combined influence of inputs on the tensile property is also simulated using ANN model. It is found that excellent performance of the ANN model was achieved, and the results showed good agreement with experimental data. Moreover, the developed ANN model can be used as a tool to control the tensile property of titanium alloys.     

Manufacturing of ultrafine-grained titanium by caliber rolling in the laboratory and in industry

The mechanical properties and the microstructure of Grade 2 titanium semi-products processed by warm caliber rolling in both laboratory and industrial environments are studied. It is shown that this technology yields ultrafine-grained (UFG) microstructure with high tensile strength and good ductility at room temperature. Finite element modelling (FEM) suggests that the effectiveness of caliber rolling in grain refinement is mainly caused by the large, homogeneous imposed strain, similar to conventional severe plastic deformation (SPD) methods. It is proved that the mechanical and microstructural properties of titanium processed by the industrial equipment are similar to the characteristics of the material manufactured in the laboratory. This observation suggests that caliber rolling carried out in industrial environments may be a candidate technology in mass-production of UFG titanium with improved mechanical properties.     

A sintering model for plasma-sprayed zirconia TBCs. Part I: Free-standing coatings

A sintering model is presented for prediction of changes in the microstructure and dimensions of free-standing, plasma-sprayed (PS) thermal barrier coatings (TBCs). It is based on the variational principle. It incorporates the main microstructural features of PS TBCs and simulates the effects of surface diffusion, grain boundary diffusion and grain growth. The model is validated by comparison with experimental data for shrinkage, surface area reduction and porosity reduction. Predicted microstructural changes are also used as input data for a previously developed thermal conductivity model. Good agreement is observed between prediction and measurement for all these characteristics. The model allows separation of the effects of coating microstructure and material properties, and captures the coupling between densifying and non-densifying mechanisms. A sensitivity analysis is presented, which highlights the importance of the initial pore architecture. Predictions indicate that the microstructural changes which give rise to (undesirable) increases in thermal conductivity and stiffness are very sensitive to surface diffusion.¹     

Through-process macro–micro finite element modeling of local loading forming of large-scale complex titanium alloy component for microstructure prediction

The intrinsic multi-heat unequal deformation behavior of the local loading forming requires a through-process macro–micro model to characterize the microstructure evolution during the forming process. In the present work, the phenomena and mechanisms of microstructural developments in local loading forming of titanium alloys are summarized. Mechanism-based unified material models, which characterize the through process microstructure evolution, are developed for integrated prediction of constitutive behavior and microstructure. A through-process macro–micro finite element model is established for the local loading forming of large scale complex titanium alloy component. The model can predict the microstructure evolution as well as macroscopic deformation in multi-step local loading forming process. Model predictions are in good agreement with experimental results. The microstructure evolution in local loading forming is investigated by the established finite element model. It is found that the thermo-mechanical processing route greatly affects the volume fraction of primary alpha but has little influence on the grain size in local loading forming     

原位钛合金化A356合金的组织与性能研究

研究了不同含钛量(分别为0.07%Ti、0.13%Ti、0.19%Ti)的原位钛合金化A356合金的微观组织和力学性能,并与相应含钛量熔配加钛A356合金进行了对比,发现原位钛合金化A356合金的二次枝晶臂间距小于熔配加钛合金,随着钛含量增加二次枝晶臂减小且趋势相同.较低钛含量原位钛合金化合金综合力学性能优于熔配加钛合金,但钛含量较高时两种合金的性能差别变小,甚至低于熔配加钛合金,原因在于较高钛含量的原位钛合金化合金中较高的铁含量严重降低了合金的塑性.     

应用人工神经网络模型预测Ti＋10V-2Fe-3A合金的力学性能

采用人工神经网络方法建立了Ti-10V-2Fe-3Al合金机械性能预测的神经网络模型。模型的输入参数包括变形温度、变形程度、固溶温度、时效温度等热加工工艺参数和热处理制度。模型的输出为钛合金最重要的5个机械性能指标，即抗拉强度、屈服强度、延伸率、断面收缩率和断裂韧性。与传统回归拟合公式相比，该模型具有容错性好、通用性强等优点。该模型可以预测Ti-10V-2Fe-3Al合金在不同热加工工艺参数和热处理制度下的机械性能，也可以用于优化热加工参数和热处理制度。