Periodic Neural Mapping for Unsteady Rotor-Blade Pressure and Aeroelastic Load Prediction

This paper introduces periodic Fourier Neural Mapping (p-FNM), a neural-operator framework designed for predicting unsteady pressure distributions on turbine rotor blades in turbomachinery applications. The method addresses the high computational cost of high-fidelity CFD simulations for aeroelastic Quantities of Interest (QoI), which depend sensitively on the temporal evolution of the pressure field. p-FNM leverages periodic Fourier features to capture the inherent periodicity in turbomachinery flows, enabling efficient prediction of blade surface pressures across different operating conditions.