SUMMARY

Metal additive manufacturing (AM) is a group of processes by which metal parts are built layer by layer from powder or wire feedstock with high-energy laser or electron beams. Metal AM can significantly improve the manufacturability of products with complex geometries and heterogeneous materials. It has the potential to be widely applied in various industries including automotive, aerospace, biomedical, energy, and other high-value low-volume manufacturing environments. However, the lack of complete process-structure-property (P-S-P) relationships for metal AM is still the bottleneck to produce defect-free, structurally sound, and reliable AM parts. There are several technical challenges in establishing P-S-P relationships for process design and optimization. First, there is a lack of fundamental understanding of the rapid solidification process during which microstructures are formed and the properties of solid parts are determined. Second, the curse of dimensionality in the process and structure design space leads to the lack of data to construct reliable P-S-P relationships.
Simulation becomes an important tool to enable us to understand the rapid solidification given the limitations of experimental techniques for in-situ measurement. In this research, we propose a mesoscale multiphysics simulation framework, called phase-field and thermal lattice Boltzmann method (PF-TLBM), with simultaneous considerations of heterogeneous nucleation, solute transport, heat transfer, and phase transition. We will develop the PF-TLBM framework to predict the microstructure evolution in the complex heating and cooling environment in the layer-by-layer AM process.
To meet the lack-of-data challenge in constructing P-S-P relationships, a new scheme of multi-fidelity physics-constrained neural network (MF-PCNN) is proposed to improve the efficiency of training in neural networks by reducing the required amount of training data and incorporating physical knowledge as constraints. Neural networks with two levels of fidelities can be combined to improve prediction accuracy. The proposed MF-PCNN will be applied to predict phase transition and dendritic growth in AM. The training of MF-PCNN is formulated as a minimax problem, and a novel training algorithm called Dual-Dimer is proposed to search high-order saddle points. A surrogate model of process-structure relationship for AM can be built based on the PF-TLBM and MF-PCNN.