About this Abstract |
Meeting |
2025 TMS Annual Meeting & Exhibition
|
Symposium
|
Algorithms Development in Materials Science and Engineering
|
Presentation Title |
Advanced Computational Techniques and Deep Learning Algorithms for the Automated Modeling and Design of Materials |
Author(s) |
Soheil Soghrati, Balavignesh Vemparala, Pengfei Zhang, Kartik Kashyap |
On-Site Speaker (Planned) |
Soheil Soghrati |
Abstract Scope |
We present an AI-driven computational framework for simulating the mechanical behavior and design of materials with complex microstructures. The first part presents an integrated microstructure reconstruction and parallel mesh generation algorithm for modeling various composite materials, including particulate, chopped fiber, and woven textile composites. In the second part, we show how this modeling framework can be used as a powerful engine for generating the training data for AL/ML applications. As an example, we show how a CNN-based model can be trained with the data generated using this framework to predict the failure response of steel pipes subjected to pitting corrosion. We also introduce a new AI-driven algorithm, Deep Learning-Driven Domain Decomposition (DLD3), that can be used as a surrogate for finite element modeling of various problems. Unlike other scientific AI/ML models, this algorithm is highly generalizable and can predict the response of problems with arbitrary geometries and loading. |
Proceedings Inclusion? |
Planned: |
Keywords |
Computational Materials Science & Engineering, Machine Learning, ICME |