A Computationally Efficient Framework for Airfoil Ice-Accretion Prediction Using Potential Flow and Lagrangian Droplet Tracking

Aircraft icing can substantially alter leading-edge geometry and degrade aerodynamic performance, highlighting the need for computationally efficient prediction methods during preliminary aircraft design. This study presents a two-dimensional reduced-order framework for airfoil ice-accretion prediction that couples a Hess–Smith potential-flow solver with Lagrangian droplet tracking, surface collection-efficiency reconstruction, a low-order freezing model, and an iterative geometry-update procedure. After each ice-accretion increment, the aerodynamic flow field and droplet trajectories are recomputed over the updated geometry, thereby capturing the coupled effects of ice growth, local flow acceleration, and downstream droplet impingement. A convergence study was performed to establish suitable surface and particle discretization. The predictive capability of the framework was assessed against four experimental NACA 23012 ice-accretion geometries representing streamwise and roughness-dominated configurations. The numerical predictions reproduced the location, extent, and principal morphological characteristics of the measured leading-edge deposits. Parametric investigations showed that the collection-efficiency distribution is governed primarily by the angle of attack, median volumetric diameter, and freestream velocity, whereas the maximum ice thickness is controlled predominantly by liquid water content and ambient temperature. Two-parameter response maps further revealed nonlinear interactions among droplet inertia, aerodynamic transport, incident water flux, freezing conditions, and geometry evolution. The proposed framework provides a practical, low-cost tool for preliminary icing assessment, sensitivity analysis, and rapid screening of atmospheric and operating conditions prior to higher-fidelity investigation.