Search

This paper proposes that an elite selection is applied to the parent population before it is directly added into the combined population, as is done in the original NSGA-II method. This modification can better avoid the algorithm's premature convergence to local optimums, and better globally explore the sample space. An analytical test problem with many local optimums is solved using the improved method. The result is compared with those of several other optimization methods. The improvements of the proposed modification are confirmed. Then the improved NSGA-II method is used to find the optimum element positions and deflection angles of the landing configuration of a multi-element airfoil, to maximize the lift coefficient. The optimal result is close to that from the original NSGA-II, while the optimum is achieved with fewer iterations. This satisfactory result shows that the improved optimization method is feasible and efficient for practical use. [ABSTRACT FROM AUTHOR]