In the NIST additive manufacturing benchmark (AM-Bench) experiments, melt pool geometry, cooling rates, surface topography, and dendritic microstructure in laser melted Inconel 625 were used to challenge and validate computational models of the melting and solidification process. To this end, three thermal models incorporating different physics are compared with the experimental data. It is identified that the heat convection enhanced by the thermocapillary flow inside the melt pool and heat loss caused by vaporization play pivotal roles to guarantee the accuracy of the predictions, and thus should be considered in the thermal model. Neglecting fluid flow and vaporization leads to nearly 100% difference in cooling rate during solidification, and 20% difference in cooling rate after solidification from the results. With the most accurate thermal model, surface topographies of the melt tracks are predicted and quantitatively analyzed. Using the Kurz-Fisher model, the primary dendrite arm spacing is predicted based on the thermal gradient and solidification rate predictions, while elemental segregation is predicted using the Scheil-Gulliver model and a non-equilibrium solidification model. Additionally, it is shown that increasing scan speed inhibits elemental microsegregation.

中文翻译：

---

Inconel 625选择性激光熔化中的热行为，表面形貌和树枝状微观结构的基准研究

在NIST增材制造基准测试（AM-Bench）中，激光熔炼Inconel 625中的熔池几何形状，冷却速率，表面形貌和树枝状微观结构被用于挑战和验证熔融和凝固过程的计算模型。为此，将三种具有不同物理性质的热模型与实验数据进行了比较。可以确定的是，熔池内部的热毛细管流增强的热对流和汽化引起的热损失在保证预测准确性方面起着关键作用，因此应在热模型中加以考虑。从结果来看，忽略流体的流动和汽化导致凝固期间冷却速率的几乎100％的差异，以及凝固后冷却速率的20％的差异。利用最精确的热模型，可以预测并定量分析熔体轨迹的表面形貌。使用Kurz-Fisher模型，基于热梯度和凝固速率的预测来预测主枝晶臂间距，而使用Scheil-Gulliver模型和非平衡凝固模型来预测元素偏析。另外，显示增加扫描速度会抑制元素微偏析。