Robust Design Optimization of Airfoils with Respect to Ice Accretion

• Published in: Journal of aircraft 2011-01, Vol.48 (1), p.287-304
• Main Authors: Ghisu, Tiziano, Jarrett, Jerome P, Parks, Geoffrey T
• Format: Article
• Language: English
• Subjects: Aerodynamics; Aerospace engineering; Air flow; Air transportation and traffic; Aircraft; Applied sciences; Design optimization; Exact sciences and technology; Freezing; Ground, air and sea transportation, marine construction; Ice
• ISSN: 0021-8669; 1533-3868
• DOI: 10.2514/1.C031100

Ice formation on an aircraft happens when supercooled water droplets impinge on the leading-edge surfaces, freezing immediately upon impact. Ice accretions alter the effective shape of the aircraft, modifying the aerodynamic forces and moments caused by the air flow over iced components. The primary effects are increased drag, reduced stall angle, and reduced maximum lift. Additional effects include decreased pitching and hinge moments that lead to a reduction in the effectiveness of control surfaces. Over the last 80 years, significant efforts have been devoted to improving understanding both of the physics of the ice accretion process and of the resulting degradation in aerodynamic performance. Significant progress has also been made through improved certification processes and more effective ice-protection systems. Despite these continued efforts, icing-induced incidents and accidents continue to occur on all classes of aircraft, to the point that the National Transportation Safety Board ranked icing among its "Most Wanted Aviation Transportation Safety Improvements." To take full advantage of these improved predictive capabilities, they need to become an active part of the design process. In this study, a computational fluid dynamics analysis of the degradation in airfoil performance caused by a quarter-round ice shape (simulating the presence of a supercooled large droplet ice accretion or runback ice) has been included in an optimization system featuring a metaheuristic multi-objective optimizer and a geometry modeler based on a freeform deformation of a datum geometry. In recognition of the real-world variability in the ice accretion location, this parameter has been treated as uncertain, with a given probability density function. An adaptive nonintrusive polynomial chaos approach has been used to propagate this input uncertainty through the otherwise deterministic system. The results presented show how, using the computational design approach developed, considerable improvements in the iced performance of the airfoil can be achieved over a range of ice locations while still maintaining good clean-wing characteristics. [PUBLICATION ABSTRACT]