We are dedicated to basic science and applied research at the intersection of materials science, manufacturing, electron microscopy, and machine learning. Our projects harness these domains to create innovative solutions that advance our understanding and capabilities in manufacturing and performance. Below are examples of our ongoing projects.
1. Real-Time Machine Learning for Microstructural Evolution in Metallic Processing
This project aims to develop a real-time machine-learning framework to monitor and predict microstructural evolution during the processing of metallic materials. By integrating in-situ characterization techniques with advanced machine learning algorithms, the framework will capture dynamic changes in microstructure as materials undergo processes like forging, rolling, or heat treatment. High-speed data acquisition systems will feed information into predictive models that can adjust processing parameters on the fly to achieve desired material properties. This approach enhances the understanding of microstructural kinetics and enables manufacturers to optimize processes in real-time, reducing defects and improving overall efficiency. The scientific contribution lies in bridging the gap between real-time data and predictive modeling, while technological impacts include more innovative manufacturing systems with adaptive control. Societally, this leads to higher-quality products and more sustainable manufacturing practices through reduced waste and energy consumption.
2. Machine Learning-Driven Design of Corrosion-Resistant Alloys
This project focuses on utilizing machine learning to accelerate the design of new metallic alloys with enhanced corrosion resistance for extreme environments. By compiling and analyzing extensive datasets on alloy compositions, microstructures, and corrosion behaviors, predictive models will be developed to identify optimal combinations of elements confer superior resistance to degradation mechanisms like pitting and stress corrosion cracking. Advanced machine learning models will handle the complex, non-linear relationships inherent in metallurgical systems, enabling the exploration of vast compositional spaces that are impractical to assess experimentally. The scientific impact includes a deeper understanding of how alloying elements influence corrosion processes, while the technological benefits involve faster development cycles for new materials. Societal advantages encompass longer-lasting infrastructure and components in marine, oil and gas, and construction industries, reducing maintenance costs and enhancing safety.
3. Multiscale Modeling and AI Integration for Mechanical Behavior Prediction in Nanostructured Metals
This project aims to develop a comprehensive multiscale modeling framework that integrates atomistic simulations with continuum mechanics and machine learning to predict the mechanical behavior of nanostructured metallic materials. The framework will capture how nanoscale features like grain boundaries and dislocations influence strength, flexibility, and toughness by bridging scales from the atomic level to the macroscopic. Machine learning algorithms will analyze simulation data, identify key patterns, and improve model accuracy. This integrative approach advances the fundamental understanding of deformation mechanisms in nanostructured metals and guides the design of materials with tailored mechanical properties. Technologically, it provides a powerful tool for engineers to predict material performance under various conditions. Societal impacts include developing advanced materials for critical aerospace, electronics, and healthcare applications, where superior performance and reliability are essential.
4. Energy-Efficient Metal Additive Manufacturing
This project focuses on applying data analytics to optimize metal processing techniques to enhance energy efficiency and reduce environmental impact. Machine learning models will identify optimal processing parameters that minimize energy consumption while maintaining or improving material properties by analyzing large datasets from processes such as casting, extrusion, and additive manufacturing. The project will also explore AI-driven predictive maintenance and process scheduling to reduce downtime and resource waste further. Scientifically, it contributes to understanding the interplay between processing conditions and energy use. Technologically, it results in intelligent manufacturing systems capable of self-optimization. For society, the project supports the transition to more sustainable industrial practices by reducing the carbon footprint of metal manufacturing, aligning with global efforts to combat climate change, and promoting responsible use of resources.
