AI assisted prediction of aircraft icing and its impacts on aerodynamic performance

About the Project

Aircraft icing may occur during the flight when supercooled water droplets within the cloud impinge on the surface where temperature is below freezing. Icing contamination on aircraft body may cause significant changes to the original aerodynamic configuration thus cause severe aerodynamics and flight mechanics degradation and threaten the flight safety seriously. Statistics show that aircraft icing is a major external cause for aircraft accidents and most of the ice-related accidents occur in the final phases of flight due to the icing contamination on wing and tailplane.

Aircraft icing at various flow conditions and flight phases has gained intensive attention with many experimental and numerical work in recent years. However, due to the dynamic and stochastic nature of ice accretion and transition of icing types over time, existing experimental and computational tests were normally conducted at specific icing conditions with limited flow variables. The expensive experimental campaigns, tremendous computational resources requirement in numerical icing simulation and the incapability of theoretical icing models to incorporate complex icing physics have caused serious issues and challenges in icing predictions and mitigation strategies. Applying deep learning-based approach to aircraft icing research could provide a promising avenue in tackling those issues.

Therefore, an innovative approach integrating artificial intelligence and statistical learning methods with the traditional experimental and computational methods is proposed in this PhD project, aiming to develop a reliable and affordable prediction tool for critical ice shapes and their impact on aircraft aerodynamic performance.

This project will require a sound understanding of aircraft aerodynamics, computational technologies (CFD), prior knowledge, experience and coding skills on machine learning techniques is desirable. Prior knowledge of aircraft icing is not necessary. It would particularly suit a graduate in Aerospace Engineering, Mechanical Engineering, or equivalent areas.

Funding Notes

there is no funding for this project