Conceptual and preliminary stages of aircraft design require significantly large number of simulations for performance analysis. As the design involves aerodynamics, structural dynamics, propulsion and controls, the number of parameters involved is considerably high. It will be time consuming to simulate various cases for analysis and design optimization. A surrogate model constructed based on few parameter sets based on Design Design of Experiments (DoEs) is quite helpful to analyse the influence of different parameters in the design space. It also helps to arrive at optimum design. The process of surrogate model development and optimization based on use of Genetic Algorithm on the surrogate model is considered in this study for the aerodynamic design of transport aircraft wing, before extending this method to include other disciplines. The Common Research Model (CRM) wing of NASA has been taken as baseline model. It has a sweep of 35°, and a maximum airfoil (t/c) of 0.1542, 0.1052 and 0.095 at root, Yehudi break and tip respectively. The design Mach number is 0.85. The design space considered in this work a) variation of Mach number from 0.75 to 0.85, b) sweep of 29° to 38°, and c) maximum t/c ratio difference of 4 % to 12% variation from baseline wing was considered in this study. The geometric model of the wing was constructed using the CATIA software. The mesh was generated with ICEM CFD software. The CFD simulation of baseline model was carried out using open domain software SU2 with RANS and Euler approaches. ANSYS Fluent software was also used to verify the results obtained from SU2 software. The surrogate surface was constructed for the three-design variable with the help of Kriging method. The influence of the design parameter on C_L, C_D and C_L/C_D was analysed. A genetic algorithm available in MATLAB was used to identify the optimum design. The predicted optimum wing configuration was again simulated to verify the confidence in the design.
展开▼
