To simulate dendrite growth with different orientations more efficiently, a high-performance cellular automata (CA) model based on heterogenous central processing unit (CPU)+ graphics processing unit (GPU) architecture has been proposed in this paper. Firstly, the decentered square algorithm (DCSA) is used to simulate the morphology of dendrite with different orientations. Secondly, parallel algorithms are proposed to take full advantage of many cores by maximizing computational parallelism. Thirdly, in order to further improve the calculation efficiency, the task scheduling scheme using multi-stream is designed to solve the waiting problem among independent tasks, improving task parallelism. Then, the present model was validated by comparing its steady dendrite tip velocity with the Lipton–Glicksman–Kurz (LGK) analytical model, which shows great agreement. Finally, it is applied to simulate the dendrite growth of the binary alloy, which proves that the present model can not only simulate the clear dendrite morphology with different orientations and secondary arms, but also show a good agreement with the in situ experiment. In addition, compared with the traditional CPU model, the speedup of this model is up to 158×, which provides a great acceleration.
Microstructure simulations of continuous casting billets are vital for understanding solidification mechanisms and optimizing process parameters. However, the commonly used CA (Cellular Automaton) model is limited by grid anisotropy, which affects the accuracy of dendrite morphology simulations. While the DCSA (Decentered Square Algorithm) reduces anisotropy, its high computational cost due to the use of fine grids and dynamic liquid/solid interface tracking hinders large-scale applications. To address this, we propose a high-performance CA-DCSA method on GPUs (Graphic Processing Units). The CA-DCSA algorithm is first refactored and implemented on a CPU–GPU heterogeneous architecture for efficient acceleration. Subsequently, key optimizations, including memory access management and warp divergence reduction, are proposed to enhance GPU utilization. Finally, simulated results are validated through industrial experiments, with relative errors of 2.5% (equiaxed crystal ratio) and 2.3% (average secondary dendrite arm spacing) in 65# steel, and 2.1% and 0.7% in 60# steel. The maximum temperature difference in 65# steel is 1.8 °C. Compared to the serial implementation, the GPU-accelerated method achieves a 1430× higher speed using two GPUs. This work has provided a powerful tool for detailed microstructure observation and process parameter optimization in continuous casting billets.
This paper studied the single crystal growth of Fe-0.64C alloy under a fixed cooling rate, proposed an improved cellular automaton model. The influence of solute field on dendrite growth is especially considered into the model. And a new solute distribution scheme based on Fick's first law and solute conservation was proposed from the microscopic scale. In addition, the present work improved the acquisition algorithm, which takes the growth length of the interface cell to the neighbouring cell as the determination condition of the capture. A solid fraction calculation method based on the solidification area of interface cell to total area is also proposed. The results show that this model can not only dynamically display the formation process of secondary dendrites and high-order dendrites, but also simulate the distribution of solute inside the dendrites and at the interface front.
Melt convection strongly influences solidification microstructure, yet its three-dimensional interaction with evolving dendritic morphology remains insufficiently quantified. Here, a three-dimensional lattice Boltzmann–cellular automata (LBM–CA) model is developed to simulate the solidification microstructure of Fe–C alloys under forced convection. The LBM resolves melt flow, and the CA model employs a decentered-octahedron capture rule to track the solid–liquid interface. The capture rule is written for a general orientation; the present calculations use only . The finite volume method solves heat and solute transport. Comparison with the Lipton–Glicksman–Kurz (LGK) solution shows that the simulated tip velocity in a quiescent melt approaches the analytical steady value. Under forced convection, the calculation is used to examine the directional difference in tip velocity, rather than as a validation against the same LGK value. The simulations reveal strong two-way coupling: forced convection thins the upstream solutal and thermal boundary layers while thickening them downstream, whereas the growing dendrite arms progressively obstruct and redistribute the surrounding flow. Increasing the inlet velocity makes this asymmetry more pronounced. At 0.0015 m/s, the upstream arm is longer than in the quiescent melt, the transverse arm is slightly longer, and the downstream arm is shorter. For the three inlet velocities examined, the downstream reduction is larger in magnitude than the upstream increase. These arm-length changes are numerical results for the present forced-convection calculations and are not compared with experimental measurements.