Constrained Aerodynamic Shape Optimisation of the Flying V Outer Wing

Aerodynamic Design of a Flying V Aircraft in Transonic Conditions

“…The only deviation is the location of the first shock at 𝑦/𝑏 = 0.80, which is predicted to be a 5% more aft compared to the experimental results. The relatively fine, structured mesh used in the Validation study also captures certain shocks, like the shock at 𝑦/𝑏 = 0.8, which were not captured in previous validation studies performed by one of the authors [17,37]. Also in comparison to the simulations of Ref.…”

“…Oosterom demonstrated that for a 20% reduction in fuel consumption with respect to the Airbus A350, the Flying V needed to have lift-to-drag ratio of at least 20.6. While Hillen and Oosterom used VLM methods for the aerodynamic optimization, Van Luijk [17] performed an optimization of the outer wing, using an Euler-based simulation in combination with a Free-Form Deformation algorithm. The parametric geometry model in ParaPy showed shortcomings in the shapes that could be generated leading to sharp transitions between sections.…”

“…The reason for this relatively low GCI lies in the fact that this research focuses on manual optimization steps, instead of having an optimizer exploring the design space. This leads to a significantly reduced number of total simulations performed and allows for stricter requirements on the accuracy of the simulation compared to previous studies [17], especially in this complicated fluid domain. As the simulation depends on several assumptions and model choices, it is important to quantify the effect these choices have on the simulation results.…”

“…Figure 12 shows the evolution of lift-to-drag ratio of the different designs. The previous parametrization, based on the work of Van Luijk [17] had an L/D of 20.3 and serves as benchmark. The L/D ratio varies from design to design, reaching a maximum of 24.2 at design 8.…”

“…Table 5 shows a summary of planform and aerodynamic characteristics of the ParaPy parametrization by Van Luijk [17] and the final design of this study. It should be noted that different methods used in both studies could lead to a slight discrepancy in values.…”