Anisotropic turbulent dispersion of particles for the prediction of total water content distribution in ice crystal icing

Conference item

Ice crystal icing, a phenomenon where atmospheric ice crystals partially melt then stick to surfaces, is a threat to jet engine operability, having been identified as the cause of numerous engine power loss and damage events. Numerical modelling of the accretion of ice relies on accurate modelling of particle transport from upstream boundary conditions, as boundary conditions for the accretion problem cannot be directly measured at surfaces. Flows relevant to ice crystal icing, both in the engine and in icing wind tunnels, are typically wall-bounded and highly turbulent. Modelling the turbulent dispersion of particles is critical to modelling particle transport in these flows, and challenging since the turbulence is inhomogeneous and anisotropic. This paper details a coupled RANS--Langevin equation approach, using only statistics sampled from a Reynolds stress equation model RANS solution of the flow field, with novel contributions to an anisotropic Langevin model. Validation is presented for a simple pipe flow case. The anisotropic model improves predictions of the distribution of particles in an altitude icing wind tunnel, the NRC RATFac, compared to experimental data and observations, versus isotropic models used in previous work.

Authors: Lenahan, M; Parker, L; Gillespie, DRH

• Publication status: Accepted
• Peer review status: Peer reviewed
• Host title: Proceedings of ASME Turbo Expo 2026
• Acceptance date: 2026-04-06
• Event title: Turbomachinery Technical Conference and Exposition 2026
• Event location: Milan, Italy
• Event website: https://event.asme.org/Turbo-Expo
• Event start date: 2026-06-15
• Event end date: 2026-06-19
• Language: English
• Keywords: computational fluid dynamics; jet engines; particulate; turbulence; multi-phase flows