Gonya Elvis M, Siyasiya Charles W, Makhatha Mamookho E
Department of Metallurgy, Faculty of Engineering and the Built Environment, University of Johannesburg, John Orr Building, Doornfontein, Johannesburg, 2028, South Africa.
Department of Materials Science and Metallurgical Engineering, University of Pretoria, Hatfield, Pretoria, 0028, South Africa.
Sci Rep. 2024 Sep 28;14(1):22496. doi: 10.1038/s41598-024-72441-9.
This work predicts, hot flow curves of 2205 DSS using strain-compensated Arrhenius rate-type constitutive model. Twenty-five (25) × Ø10 diameter × 15 mm height cylindrical samples were hot compressed at a temperature between 850 and 1050 °C at an interval of 50 °C and strain rates between 0.001 and 5 s, using Gleeble 1500D. After the tests, corrected flow curves were plotted followed by computation of deformations constants at various deformation conditions using steady state stress. The values of the constants were (α = 0.009708, Q = 445 kJ/mol and n = 3.7) and seemed comparable to the previous studies of DSS. Steady state predictive model was then constructed using the calculated constants and showed a reasonably good accuracy with low value of MARE = 7.78%. Furthermore, calculated strain compensated Arrhenius rate type model was used to predict flow curves at various deformation. The model had a good estimation of flow curves of flow curves at 900-1050 °C across all strain rates as reflected by MARE = 5.47%. A notable discrepancy between predicted and experimental flow stress was observed at 850 °C and across all the strain rates. A model refinement using generalised reduced gradient improved the accuracy of the model by 34.7% despite deformation conditions at 850 °C and low strain rates (0.01/ 0.1) s showing minimum improvement. Further modification of Z-parameter by compensating for the strain rate improved the accuracy of the model at 850 °C/0.01 s/0.1 s. Lastly, a comparison of the current model with the other non-linear model showed that the latter was more accurate in estimation of flow curves since it relied on characteristics flow stress points controlled by underlying active deformation mechanisms.
本研究采用应变补偿阿伦尼乌斯速率型本构模型预测了2205双相不锈钢的热加工曲线。使用Gleeble 1500D对25个直径为Ø10、高度为15mm的圆柱形试样在850至1050°C的温度范围内以50°C的间隔以及0.001至5s的应变速率进行热压缩。试验后，绘制校正后的流动曲线，然后使用稳态应力计算各种变形条件下的变形常数。常数的值为(α = 0.009708, Q = 445kJ/mol，n = 3.7)，与之前对双相不锈钢的研究结果相当。然后使用计算得到的常数构建稳态预测模型，该模型显示出相当好的精度，平均绝对相对误差(MARE)值较低，为7.78%。此外，使用计算得到的应变补偿阿伦尼乌斯速率型模型预测各种变形下的流动曲线。该模型对900 - 1050°C所有应变速率下的流动曲线具有良好的估计能力，MARE = 5.47%。在850°C以及所有应变速率下，预测的流动应力与实验值之间存在显著差异。尽管在850°C和低应变速率(0.01/0.1)s下变形条件的改善最小，但使用广义简约梯度法对模型进行优化后，模型精度提高了34.7%。通过补偿应变速率对Z参数进行进一步修正，提高了模型在850°C/0.01s/0.1s下的精度。最后，将当前模型与其他非线性模型进行比较，结果表明后者在流动曲线估计方面更准确，因为它依赖于由潜在的主动变形机制控制的特征流动应力点。
