Mahmood Muhammad Arif, Ur Rehman Asif, Khraisheh Marwan
Mechanical Engineering Program, Texas A&M University at Qatar, Doha, Qatar.
ERMAKSAN, Bursa, Turkey.
3D Print Addit Manuf. 2024 Jun 18;11(3):e1366-e1379. doi: 10.1089/3dp.2023.0016. eCollection 2024 Jun.
In this work, we propose a methodology to develop printability maps for the laser powder bed fusion of AISI 316L stainless steel. Regions in the process space associated with different defect types, including lack of fusion, balling, and keyhole formation, have been considered as a melt pool geometry function, determined using a finite element method model containing temperature-dependent thermophysical properties. Experiments were performed to validate the printability maps, showing a reliable correlation between experiments and simulations. The validated simulation model was then applied to collect the data by varying laser scanning speed, laser power, powder layer thickness, and powder bed preheating temperature. Following this, the collected data were used to train and test the adaptive neuro-fuzzy interference system (ANFIS)-based machine learning model. The validated ANFIS model was used to develop printability maps by correlating the melt pool characteristics to the defect types. The smart printability maps produced by the proposed methodology can be used to identify the processing window to attain defects-free components, thus attaining dense parts.
在这项工作中，我们提出了一种为AISI 316L不锈钢激光粉末床熔融开发可打印性地图的方法。与不同缺陷类型相关的工艺空间区域，包括未熔合、球化和匙孔形成，已被视为熔池几何形状的函数，使用包含温度相关热物理性质的有限元方法模型来确定。进行了实验以验证可打印性地图，显示出实验与模拟之间可靠的相关性。然后应用经过验证的模拟模型，通过改变激光扫描速度、激光功率、粉末层厚度和粉末床预热温度来收集数据。在此之后，收集到的数据用于训练和测试基于自适应神经模糊推理系统（ANFIS）的机器学习模型。经过验证的ANFIS模型通过将熔池特征与缺陷类型相关联来开发可打印性地图。所提出的方法生成的智能可打印性地图可用于识别加工窗口，以获得无缺陷部件，从而获得致密零件。
