Computationally Efficient Algorithm for Modeling Grain Growth Using Hillert's Mean-Field Approach

Area of Science:

• Materials Science
• Computational Materials Science
• Metallurgy

Background:

• Understanding microstructure evolution during hot-rolling is crucial for optimizing material properties.
• Existing models often struggle with computational efficiency for large-scale simulations and long durations.
• Accurate modeling requires capturing complex interactions between individual grains.

Purpose of the Study:

• To develop a novel, efficient through-process model for simulating microstructure evolution during hot-rolling.
• To accurately describe grain growth in larger systems and over extended simulation times.
• To provide a versatile tool for investigating the effects of manufacturing processes on material microstructure.

Main Methods:

• Implementation of a novel numerical mean-field approach where each grain interacts with an average medium.
• Utilization of an upsampling approach to dynamically adjust the simulation grain ensemble for efficiency and accuracy.
• Verification of the model against analytical solutions and experimental data.

Main Results:

• The mean-field model efficiently simulates grain growth, avoiding complexities of individual grain interactions.
• The upsampling approach prevents undersampling artifacts and ensures accuracy regardless of initial grain count.
• Model accuracy is validated through high agreement with analytical solutions and experimental data.
• Successful investigation of different initial conditions demonstrates the model's versatility.

Conclusions:

• The developed mean-field model offers a simple and efficient solution for simulating microstructure evolution during hot-rolling.
• The model's accuracy and adaptability make it suitable for investigating manufacturing process effects.
• Its seamless integration capability allows for broader application in microstructure evolution studies.