02.4 - Applied AerodynamicsAERODYNAMIC SHAPE OPTIMIZATION WITH A NEW PARALLEL EVOLUTIONARY ALGORITHM AND NUMERICAL FLOW MODELLINGA. Jahangirianš, M. Ebrahimiš; šAmirkabir University of Technology, Iran A hybrid Parallel Genetic Algorithm (PGA) that combines a multi-population PGA and masterslave is applied to accelerate aerodynamic shape optimization. The method is based on an adaptive method that updates GA operators in each generation by taking feedback from the current population. An unstructured grid finite volume flow solver is utilized for the objective function evaluations. The performance of the method is further evaluated by an aerodynamic shape optimization. Results indicate that not only the maximum objective value is improved but also the total computational time is decreased significantly. |