What are you trying to do? Articulate your objectives using absolutely no jargon.
In our lab, we try to predict when, where, why, and how structural materials fail, with an emphasis on fatigue and fracture.
How is it done today, and what are the limits of current practice?
Most predictions of material failure fall short in terms of representing the salient microstructural features that play a role in mechanical behavior, or, if they do succeed in representing the salient features, are too computationally expensive to solve meaningful problems in real applications.
What is new in your approach and why do you think it will be successful?
We implement and leverage state-of-the-art computational frameworks to represent the salient microstructural features that play governing roles in material deformation and failure, while pushing the limits of computational efficiency. Our physics-based models include crystal plasticity implementations with novel approaches for capturing grain-size effects, representing three-dimensional fracture surfaces, and performing high-throughput microstructural simulations. Our approach to data science leverages our domain expertise to account for the nuances of feature engineering, data sampling, cross validation, and requisite context needed to achieve an envisioned application of AI.
Who cares? If you are successful, what difference will it make?
The physics-based and AI models that we develop will enable advanced structural prognosis as well as materials design for a broad range of applications.