A GPU-accelerated 3D PF-LBM modelling of multi-dendritic growth in an undercooled melt of Fe–C binary alloy

• DOI: 10.1016/j.jmrt.2022.01.132
• Publication Date: 2022-02-02T01:47:03Z
• Authors: Yaqi Guo; Sen Luo; Weiling Wang; Miaoyong Zhu
• Keywords: Equiaxed crystals; Lattice Boltzmann methods; Dendrite (mathematics)

In the present study, a GPU-accelerated 3D PF-LBM model, which is combination of phase field method (PF) and lattice Boltzmann (LBM), established to predict multi-dendritic growth Fe–C binary alloy. order solve problems insufficient memory low data transfer efficiency faced by GPU-based parallel computation for dendritic growth, segment loading implemented with cudaMemcpy function in computing architecture CUDA, optimum thread block size 128 determined. The performance analysis model shows that gets rid limitation GPU speedup ratio reaches 1700 times cell number 5123. Then, used investigate stationary single/multiple an undercooled melt Fe-2mol pct C alloy or without forced fluid flow 0.05 m/s, results show rejected solute at solid/liquid interface washed away enriched downstream. But enrichment interdendritic space hardly flow, when dendrites are fully developed. Thus, becomes more significant downstream region, especially space. Therefore, equiaxed dendrite morphology asymmetrical multiple-dendritic flow.