SBIR/STTR Award attributes
Ice accretion on aircraft can trigger flow separation and degrade aerodynamic performance by reducing lift and stall angle-of-attack, increasing drag, and in severe cases causing complete loss of aircraft control. Modeling and quantification of icing effects on aircraft performance therefore plays a critical role in aircraft design and certification. High-fidelity CFD analyses of aircraft with imposed ice shapes are impeded by time-consuming manual pre-processing and mesh generation that are difficult to automate. Given the desire to adapt the mesh to optimize a given output functional (e.g. aircraft CLmax), these challenges are particularly important as the mesh quality in the vicinity of complex ice shapes directly impacts the accuracy of CFD solution error estimates. The objective of this project is to develop, demonstrate, and deliver a high-resolution automated unstructured mesh refinement framework for aircraft icing predictions. The capability will interface with existing NASA CFD solvers and provide access to high-resolution icing data in a manner consistent with established procedures for accessing CAD geometry, while locally disambiguating between CAD and ice shape. Grid quality improvements will be made near the complex ice geometry to improve error estimates. In Phase I, the capability will be developed in FUN3D using metric-based anisotropic mesh refinement to achieve optimal CLmax prediction for an iced aircraft configuration. Mesh refinement at the ice surface will be augmented to leverage the new API to query the true ice shape to improve the resolution of the surface discretization. Accuracy and efficiency of the developed capability will be demonstrated for a canonical wing geometry as proof of concept. Phase II efforts will further develop and mature the capability, and demonstrate on more complex topologies including high lift geometries and perform uncertainty quantification to understand and improve solution sensitivity and accuracy.
