The hot deformation behavior of Ti2AlNb-based alloys in the temperature range of 650~850 ℃ and strain rate range of 0.001~1 s-1 was investigated by isothermal constant strain rate compression tests, and the three-dimensional processing maps were constructed based on dynamic material modelling. The flow stress curves of Ti2AlNb-based alloys were analyzed and the SVM constitutive model was established, then the 3D processing maps were theoretically analyzed, and finally the accuracy of the constructed 3D processing maps was verified with the microstructures. The results show that the flow stress of Ti2AlNb-based alloys increases with the decreasing of deformation temperature and the increasing of strain rate. The SVM model may accurately predict the flow behavior of Ti2AlNb-based alloys under different deformation processing parameter conditions, with a correlation factor of 0.999 and an average relative error of 0.67%. The 3D processing maps show that the regions with high values of power dissipation efficiency η are concentrated in the region of low strain rate. The better hot deformation processing parameters for Ti2AlNb-based alloys under different strains are as 675~725 ℃, 0.001~0.003 s-1, and the optimal hot deformation processing parameters are as 700 ℃, 0.001 s-1.
对于SVM模型的构建,假设有n个原始数据 xi (i=1,2,…,n),每个数据对应一个目标yi (i=1,2,…,n)。组合后成为一个由n维向量{( xi,yi ),i=1,2,…,n}组成的样本集,其中 xi ∈Rn 为输入变量(热压缩实验参数:真应变ε、变形温度T、应变速率),yi ∈R为输出变量(流动应力σ)。通过将原本非线性的分类问题转化为线性可分的问题,得到最优回归函数f( x )=ωφ( x )+b,其中 ω 为空间权重因子矢量,φ( x )为样本点 x 的目标函数,b为偏差。根据结构最小化原则,目标函数和约束条件设置如下[20]:
PENGZhenlong, ZHANGXiangyu, LIULiangbao, et al. Effect of High-speed Ultrasonic Vibration Cutting on the Microstructure, Surface Integrity, and Wear Behavior of Titanium Alloy[J]. Journal of Materials Research and Technology2023, 24: 3807-3888.
CAOZiwen, ZHUMing, PUZhineng, et al. Effect of Sintering Temperature on Microstructure and High Temperature Tensile Properties of Ti2AlNb Alloy[J]. Transactions of Materials and Heat Treatment,2024, 45(8): 77-84.
CEHNJinkun, JIANGFengyang, SIFang, et al. Effect of Second-sintering Temperature on Microstructure and Mechanical Properties of Ti2AlNb Prepared by Spark Plasma Sintering[J]. Rare Metals and Cemented Carbides, 2024, 52(4):54-58.
MAXiuping, PUZhiqiang, LIRan, et al. Microstructure and Mechanical Properties of Thermal Simulation Welding Heat-affected Zone of Ti2AlNb Alloy[J]. Transactions of Materials and Heat Treatment, 2024, 45(2): 204-210.
WULiang, CHENZheng. Study on SVM Based Mathematical Model Used to Predict Mechanical Properties of Materials after Heat Treatment[J]. Transactions of Materials and Heat Treatment, 2007, 59(6): 152-155.
[12]
LIULimeng, CHUMaoxiang, GONGRongfeng, et al. Unbalanced Classification Method Using Least Squares Support Vector Machine with Sparse Strategy for Steel Surface Defects with Label Noise[J]. Journal of Iron and Steel Research International, 2020, 27: 1407-1419.
YINXiaoshan, ZHONGJianlin, PENGBaoying,et al. Prediction of Ti2AlNb Cutting Force Based on Sparrow Algorithm Optimized Least Squares Support Vector Machine[J]. Tool Engineering, 2023, 57(10): 63-68.
[15]
SIMK H, RIY C, JOC H, et al. Modified Zerilli-Armstrong and Khan-Huang-Liang Constitutive Models to Predict Hot Deformation Behavior in a Powder Metallurgy Ti-22Al-25Nb Alloy[J].Vacuum, 2023, 210: 111749.
[16]
CHAIYunpeng, ZHUYanchun, QINLing, et al. High-temperature Hot Deformation Behavior and Processing Map of Ti-22Al-25Nb Alloy[J]. Materials Today Communications, 2024, 41: 110599.
[17]
PRASADY V R K, GEGELH L, DORAIVELUS M, et al. Modeling of Dynamic Material Behavior in Hot Deformation: Forging of Ti-6242[J]. Metallurgical Transactions A,1984, 15(10): 1883-1892.
LIUXiaoyan, ZHANGXiwei, CHENXiuquan, et al. Hot Deformation Behavior and Deformation Mechanism of Hot Extruded FGH96 Superalloy[J]. Chinese Journal of Rare Metals, 2024, 48(8): 1108-1119.
[20]
LIChunhong, QIURisheng, LUANBaifeng, et al. Hot Deformation and Processing Maps of As-sintered CNT/Al-Cu Composites Fabricated by Flake Powder Metallurgy[J]. Transactions of Nonferrous Metals Society of China, 2018, 28(9): 1695-1704.
ZHOUWei, GUANFengtao, QIANXin'an, et al. Application of Artificial Neural Network in Establishment of Constitutive Relationship Model of Coarse Grained Ti2AlNb[J]. Hot Working Technology, 2023, 52(22): 107-109.
[23]
LIANGShengli, ZHOUMeng, ZHANGYi, et al. Thermal Deformation Behavior of GO/CeO2 In-situ Reinforced Cu30Cr10W Electrical Contact Material[J]. Journal of Alloys and Compounds, 2022, 899: 163266.
[24]
RANJANA, MAHARAJAH, MISHRAS, et al. Microstructural Effects and Constitutive Modelling of Cyclic Softening Behaviour in Ti-6Al-4V Titanium Alloys[J]. Materials Science and Engineering A, 2024, 901: 146527.
ZHANGZhixiong, ZHANGJuntao, HANJian-chao, et al. Microstructure and Mechanics Property Variations Duringmdlf of Ti-6Al-4V Alloy with a Martensitic Microstructure[J]. China Mechanical Engineering, 2021, 32(22): 2739-2748.
[27]
GAOWei, XUFan, ZHOUZhihua. Towards Convergence Rate Analysis of Random Forests for Classification[J]. Artificial Intelligence, 2022, 313: 103788.
[28]
ZHANGTong. Statistical Behavior and Consistency of Classification Methods Based on Convex Risk Minimization[J]. The Annals of Statistics, 2004, 32(1): 56-85.
[29]
李航. 统计学习方法[M]. 北京: 清华大学出版社, 2012.
[30]
LIHang. Statistical Learning Method[M]. Beijing: Tsinghua University Press, 2012.
[31]
TAKHANOVR. On the Speed of Uniform Convergence in Mercer's Theorem[J]. Journal of Mathematical Analysis and Applications, 2023, 518(2): 126718.
[32]
PANBinbin, LAIJianhuang, CHENWensheng. Nonlinear Nonnegative Matrix Factorization Based on Mercer Kernel Construction[J]. Pattern Recognition, 2011, 44(10/11): 2800-2810.
[33]
SUYKENSJ A K, VANDEWALLEJ. Least Squares Support Vector Machine Classifiers[J]. Neural Processing Letters, 1999, 9: 293-300.
[34]
WEND X, LINY C, LIH B, et al. Hot Deformation Behavior and Processing Map of a Typical Ni-based Superalloy[J]. Materials Science and Engineering A, 2014, 591: 183-192.
[35]
MURTYS V S. N, RAO B N. On the Development of Instability Criteria during Hotworking with Reference to IN 718 [J]. Materials Science and Engineering: A, 1998, 254(1/2): 76-82.
[36]
DONGXuemao, XUJing, FENGZhongxue, et al. Exploring Hot Deformation Behavior of the Solutionized Cu-15Ni-8Sn Alloy through Constitutive Equations and Processing Maps[J]. Journal of Materials Research and Technology, 2024, 29: 2142-2153.
JIANGSenbao, WANGYusheng, CHENYao, et al. Laser Bending Process and Microstructure for Ti2AlNb Titanium Alloy Rolled Sheet[J]. Forging & Stamping Technology, 2024, 49(5): 61-66.
[39]
WURonghai, LIUYang, GENGCong, et al. Study on Hot Deformation Behavior and Intrinsic Workability of 6063 Aluminum Alloys Using 3D Processing Map[J]. Journal of Alloys and Compounds, 2017, 713: 212-221.
[40]
KONGYonghua, CHANGPengpeng, LIQian, et al. Hot Deformation Characteristics and Processing Map of Nickel-based C276 Superalloy[J]. Journal of Alloys and Compounds, 2015, 622: 738-744.
[41]
LIChunxiao, JINJianfeng, YANHong, et al. Investigation of Hot Deformation Behavior of Mg-14Gd-0.5 Zr (wt%) through Constitutive Analysis and Processing Maps[J]. Journal of Alloys and Compounds, 2023, 953: 170112.
CAIShuqiao, WANGYongfan, CHENKexin, et al. Hot Deformation Behavior and Hot Processing Map of 18Cr2Ni4W Steel[J]. Transactions of Materials and Heat Treatment, 2024,45(7): 201-211.
WANGBin, ZHANGKaifeng, JIANGShaosong, et al. Effect of Solution Treatment Temperature on Microstructural Evolution of Ti2AlNb-based Alloy[J]. Journal of Aeronautical Materials, 2015, 35(3): 7-12.