论文部分内容阅读
为研究TC4钛合金的力学性能,借助Gleeble-1500D热模拟机,在等温、等速率条件下,对Ti-6Al-4V合金进行压缩。为进一步分析实验结果,将非线性建模方法高斯回归过程(GP)应用于TC4流变应力的预测,并提出相应的模型。实例研究表明,借助MATLAB语言编制程序,预测流变应力的高斯回归过程学习方法是科学、可行的,其绝对误差为0.91MPa,相对误差为4.84%,与ANN模型相比,预测精度更高,而且简单实用,为TC4热变形工艺的制定提供参考依据。
In order to study the mechanical properties of TC4 titanium alloy, the Ti-6Al-4V alloy was compressed under the condition of isothermal and constant velocity by Gleeble-1500D thermal simulator. In order to further analyze the experimental results, the Gauss Regression Process (GP), a nonlinear modeling method, was applied to the prediction of TC4 flow stress and the corresponding model was proposed. The case study shows that it is scientific and feasible to study the Gaussian regression process of predicting the flow stress with MATLAB language programming. The absolute error is 0.91 MPa and the relative error is 4.84%. Compared with the ANN model, the prediction accuracy is higher, But also simple and practical, providing reference for the formulation of TC4 thermal deformation process.