Browsing by Author "Palmeira, Ricardo Jorge Faria"
Now showing 1 - 1 of 1
Results Per Page
Sort Options
- Low speed aerofoil optimizationPublication . Palmeira, Ricardo Jorge Faria; Gamboa, Pedro VieiraAerofoil shape has a significant influence on aircraft performance. Multiple methodologies can be applied, such as direct design, inverse design or performance design. With the improvement of computer technology there has been a continuing trend of automating this process by using performancebased methods and formal optimisation algorithms. Parametrization formulations of aerofoils have continually advanced, some examples are BSpline, Class Shape Functions, HicksHenne functions and BezierPARSEC 3333. Main comparisons of parametrizations have focussed on morphology, design space and aerodynamic consistency. In the present work, the parametrizations mentioned are applied to aerofoil optimisation and their results compared for different numbers of design variables, in order to ascertain optimisation differences. Performance design optimisation is used in a multipoint approach with an aggregated objective function using weights that are determined using the aircraft design data, to maximize the score for the competition Air Cargo Challenge (ACC2019 and ACC2022), using XFOIL for aerodynamic analysis and particle swarm optimisation (PSO) under a modified version of the XOPTFOIL tool. The initial aerofoil was obtained by iterative inverse design during previous works, the optimisation includes the flap chord and deflection angle for the different selected lift coefficient conditions as design variables. The initial population is bounded between maximum and minimum limits set by the initial aerofoil design variables and an initial perturbation. The aerofoil is constrained by minimum and maximum thicknesses, a minimum trailing edge angle and a specified trailing edge thickness. Several additional restrictions are also imposed on the aerofoil to avoid unneeded analysis of a geometry with an expected non converged solution in XFOIL. These include the angles’ maximum, minimum and difference values of the two points closest to the leading edge, the maximum angle between any three consecutive points and the number of curvature sign reversals at the upper surface and lower surface of the aerofoil. To deal with the constraints and restrictions a penalty function is used, each penalty being normalised by a maximum set value. To ensure that these do not unduly constrain the domain exploration of the optimisation, a dynamic limit to the penalties is used. During the optimisation, this limit decreases linearly with the iterations. From two case studies, it was possible to demonstrate the tool ability to optimize aerofoils. In the first case, utilisation of BSplines achieved better results relative to the other methods. In the second case, the dynamic limit, consistency method and XFOIL convergence recuperation method are studied. This last one has the greatest influence on optimisation.