This thesis focuses on the study of precipitation kinetics in Al–Zn–Mg–Cu alloys, widely used in the aerospace industry for their high mechanical performance. The evolution of precipitates governs the in-service properties. Atom probe tomography revealed the non-stoichiometry of precipitates and a copper enrichment in the smallest ones.
A high-throughput characterization methodology, combining in situ small-angle X-ray scattering with composition- and process-graded specimens, enabled the study of the effects of composition, aging, quenching, and plastic deformation on precipitation. Based on these findings, a non-stoichiometric physical model and its machine learning surrogate were developed. The latter, one million times faster, predicts microstructural evolution and mechanical properties, paving the way for optimized alloy and process design in industrial applications.