High-fidelity and data-driven approach to evaluate wind effects on structures
Please login to view abstract download link
Wind–structure interaction is difficult to model due to its unsteady and multi-physics nature. This work presents a fast, AI-assisted framework that integrates wind tunnel data and numerical simulations to predict wind-induced structural responses. Wind loading is modeled using a Koopman operator, while structural behavior is evaluated with classical finite element methods, enabling efficient and reliable assessment of structures under wind loads without resorting to computationally expensive fully coupled simulations.
