基于PSO-LSSVM的Ni-Cr-W-Mo合金本构关系研究  

Research on constitutive relation of Ni-Cr-W-Mo alloy based on PSO-LSSVM

作　　者：蔡改贫[1] 罗茜茜 刘鑫 CAI Gai-pin;LUO Xi-xi;LIU Xin(School of Mechanical and Electrical Engineering,Jiangxi University of Science and Technology,Ganzhou 341000,China)

机构地区：[1]江西理工大学机电工程学院,江西赣州341000

出　　处：《塑性工程学报》2020年第9期140-146,共7页Journal of Plasticity Engineering

基　　金：国家自然科学基金资助项目(51905241);江西省教育厅科学技术研究项目(GJJ170560)。

摘　　要：针对热成形金属材料变形本构关系的多因素、高度非线性的特点,利用Gleeble-3500型热模拟试验机对Ni-CrW-Mo镍基高温合金开展变形温度和应变速率分别为950~1200℃和0.01~10 s^-1的平面应变热压缩实验。建立材料本构关系的LSSVM模型,并采用粒子群算法(Particle Swarm Optimization,PSO)对LSSVM模型的惩罚因子c和核宽度δ进行寻优,构造了基于PSO-LSSVM的Ni-Cr-W-Mo镍基高温合金本构模型。将实验数据与模型预测值进行对比,结果表明,所建本构模型预测值与实验值间的平均相对误差仅为1.98%,模型预测精度高、泛化能力强,能够准确预测Ni-Cr-W-Mo镍基高温合金的高温流动应力。Aiming at the multi-factor and highly non-linear characteristics of deformation constitutive relation of hot-formed metal materials,the plane strain thermal compression experiments of Ni-Cr-W-Mo nickel-based superalloy were carried out on Gleeble-3500 thermal simulator with the deformation temperature of 950-1200℃and the strain rate of 0.01-10 s^-1.Least Squares Support Vector Machine(LSSVM)model of constitutive relation of the material was established,and particle swarm optimization(PSO)was applied to optimize the regularization factor c and kernel widthδof LSSVM model,the constitutive model of Ni-Cr-W-Mo nickelbase superalloy based on PSO-LSSVM was constructed.Comparing the experimental data with the predicted values of the model,the results show that the average relative error between the predicted values and experimental values is only 1.98%.The model has high prediction accuracy and strong generalization ability,which can accurately predict the high temperature flow stress of Ni-Cr-W-Mo nickel-based superalloy.

关 键 词：本构模型 PSO-LSSVM 镍基合金 平面应变热压缩实验 

分 类 号：TG146.15[一般工业技术—材料科学与工程] TP181[金属学及工艺—金属材料]