| Abstract Scope |
Additive friction stir deposition is a solid-state additive process for high-integrity repair and build-up of structural components without melting. It is particularly suited for aerospace and defense applications (e.g., repairing deep volumetric damage), producing fully dense deposits with markedly transformed microstructures while avoiding fusion-related defects such as porosity and hot cracking. This work establishes a unified framework integrating in situ monitoring, process modeling, and AI to uncover governing process physics and enable performance optimization. A comprehensive sensing platform captures temperature, force, torque, heat flux, and material flow, while physics-based modeling and Bayesian learning fuse such data into predictive, interpretable process descriptors. This integration resolves three-dimensional material flow, voxel-level thermomechanical histories, and shear-driven transport mechanisms governing bonding and microstructure, establishing quantitative process–structure–property linkages. This capability is demonstrated through a representative case of additive repair of aluminum fastener holes in aerospace applications, where post-repair fatigue performance exceeds the pristine condition. |