Abstract

The thermal compression tests of Ti-22Al-26Nb-2Ta alloy under T = 1173 ~ 1423 K and \(\dot{\varepsilon }\) = 0.001 ~ 10 s⁻¹ were carried out on the Gleeble-3500 thermo-mechanical simulator. The flow stress curves were obtained, and the high-temperature rheological properties of the alloy were analyzed. The 3D activation energy maps were calculated and constructed. The least squares support vector machine (LSSVM) model of constitutive relation was established, and the penalty coefficient and kernel parameter of the LSSVM model were optimized by genetic algorithm (GA). The constitutive model of the alloy based on the GA-LSSVM algorithm was constructed. The predicted value of the model was also compared with the experimental data. The dynamic material model (DMM) and polar reciprocity model (PRM) were used to establish the 3D processing map of the alloy and appropriate thermal processing parameters. Our researches indicated that deformation temperature and strain rate have a great influence on the flow stress of Ti-22Al-26Nb-2Ta alloy. Ti-22Al-26Nb-2Ta alloy is a negative temperature-sensitive and a positive strain rate-sensitive material. The correlation coefficient of GA-LSSVM algorithm constitutive model is 0.9922, and the relative error of most samples is within 10%, accounting for 93.18%. The model has high prediction accuracy and strong generalization ability. The DMM processing map based on the Prasad instability criterion is more accurate in optimizing the processing parameters of the alloy than that of the PRM processing map through analyzing the 3D processing map and observing the microstructure. The instability modes in the instability region of the alloy mainly include adiabatic shear, crack, and local flow. The 1173 ~ 1273 K/0.001 ~ 0.003 s⁻¹ are the best parameters during the processing of the alloy.

Graphic Abstract

中文翻译：

---

基于GA-LSSVM和3D工艺图的Ti-22Al-26Nb-2Ta合金的热变形行为

摘要

Ti-22Al-26Nb-2Ta合金在T  = 1173〜1423 K和\（\ dot {\ varepsilon} \）  = 0.001〜10 s ^-1时的热压缩试验在Gleeble-3500热机械模拟器上进行。获得了流动应力曲线，并分析了合金的高温流变性能。计算并构建了3D激活能图。建立了本构关系的最小二乘支持向量机（LSSVM）模型，并通过遗传算法（GA）对LSSVM模型的惩罚系数和核参数进行了优化。建立了基于GA-LSSVM算法的合金本构模型。模型的预测值也与实验数据进行了比较。动态材料模型（DMM）和极易互易模型（PRM）用于建立合金的3D处理图和适当的热处理参数。我们的研究表明，变形温度和应变速率对Ti-22Al-26Nb-2Ta合金的流变应力有很大的影响。Ti-22Al-26Nb-2Ta合金是负温度敏感性和正应变率敏感性材料。GA-LSSVM算法本构模型的相关系数为0.9922，多数样本的相对误差在10％以内，占93.18％。该模型具有较高的预测精度和较强的泛化能力。通过分析3D加工图并观察显微组织，基于Prasad不稳定性准则的DMM加工图在优化合金加工参数方面比PRM加工图更准确。合金失稳区域的失稳模式主要包括绝热剪切，裂纹和局部流动。1173〜1273 K / 0。^-1是合金加工过程中的最佳参数。

图形摘要