Abstract Scope |
We present an overview of existing and emerging machine learning (ML) applications in the design, synthesis, and characterization of metal matrix composites (MMC). We have shown that machine learning approaches can be used in three different categories: property prediction, microstructure analysis, and process optimization, which are correlated with three different types of machine learning techniques: regression, classification, and optimal control, respectively. Mechanical, tribological, corrosion, and wetting properties of various MMCs have all been successfully predicted using machine learning algorithms. However, despite their enormous capabilities, ML methods such as computer vision, which is useful for microstructural characterization and defect detection, and optimization algorithms (e.g., reinforcement learning) have not been widely utilized for the design, processing, and characterization of metal matrix composites. We conclude that ML offers enormous opportunities to gain more knowledge about MMC’s; they can help design, manufacture, and deploy new MMC’s significantly faster at a fraction of the cost. |