ONR

Duration: 2024-2027

Role: PI  

State-of-the-art Digital-Twin for real time assessment of structural health and estimating the reliability of the structural operations is far from its realization. Digital-Twin is envisioned in a way where it should be able to perform the real time assessment of the responses from the existing and new naval structures. Visualization, estimation, and prediction of the structural health of the naval structures will be critical for the future manned or unmanned naval fleets through integrated sensory system coupled with their Digital-Twin. Physics based understanding of the sensory data from new generation of material systems with existing and new sensor system is utmost important for seam less functionality of the Digital-Twin. However, due to several bottle neck challenges, this goal is obscured. Currently no physics-based link exists with the Digital-Twins of Naval assets that can derive intelligent decision form the sensor data in real time. Hence, the project objective is to create a universal physics-deep-link for the Digital Twins of naval assets with existing and new material systems. We propose a close collaboration and continuous interaction with NSWC Carderock’s material and SHM team with our undergraduate and graduate students to create a universal Physics Deep link interface that could easily interface with Navy’s current Digital Twin software. The proposed physics deep-link will be developed through extensive computational and experimental study that will house trained machine learning (ML) models and artificial intelligence (AI) for real time damage estimation and prediction in naval structures. While contributing to the scientific merit we propose to achieve additional broader impact through training and education of our students using a unique Expose, Engage, Educate and Evaluation (E4) model towards the development of workforces for the future Navy.