Aerodynamic Shape Optimization of Multi-Element Airfoils in Ground Effect

Abstract

Relatively little published research addresses wings in ground effect with a moving ground plane. Due to the important effect of the ground, fixed ground tests are not believed to be reliable indicators of either performance or critical physical phenomena, and may not be any more useful than free-air data. One method to model the ground for experimental studies is tangential blowing—injecting flow close to the ground at the freestream velocity; however this approach is not widely adopted, due to the complexity and accuracy of the system relative to that of the physically correct moving-belt ground plane. The method of images uses a second geometry, inverted and placed at double the ground distance below the first. This technique has been employed in wind tunnel tests, and computational panel methods, but does not maintain an accurate ground-plane in all conditions. Previous computational studies using Reynolds Averaged Navier-Stokes (RANS) solvers include, and the importance of modeling the ground plane as moving with the freestream velocity as compared to other boundary conditions is demonstrated and discussed in. The first successful tests using a moving belt were conducted in the 1930’s, and are now common practice for ground vehicle studies. For airfoils with negative lift coefficients, the studies with a moving ground plane indicate that the general behavior is such that as the distance between the wing and ground is reduced, the lift coefficient is first amplified, and as the gap becomes very small, the lift force then reduces. The design of high-lift systems can be very complicated due to the number of design parameters as well as the multiple aerodynamic phenomena and interactions (e. g., merging boundary layers and wakes, transition to turbulence, separation), and is typically accomplished with detailed wind tunnel testing. More recently, CFD analyses have been incorporated in the design process. Design optimizations have also been attempted. For example, Eyi et al. employed the incompressible Navier-Stokes equations and a chimera overlaid grid system, and Besnard et al. used an Interactive Boundary Layer (IBL) approach. The performance gradients in these earlier works were obtained through finite-difference methods, and thus were only able to span a small design space (i. e., the rigging parameters) due to the large computational cost associated with the finite-difference approach. Parallel to this study, a control theory approach using the Navier-Stokes equations was independently developed by Kim, Alonso, and Jameson. Their work included