Abstract:

Based on the definition of Pareto optimum, a Pareto Genetic Algorithm (PGA) suitable for multi objective optimization is developed using a set of operators and strategies in this paper. The search process of the PGA is speeded up by presenting a ranking method. The distribution of the solutions on the Pareto front is improved from two angles. Firstly, the niching technology is used to keep the genetic diversity. Secondly, the real coding combined with an average crossover operator is used to eliminate the ending effect of the GA coding. Then the constructed PGA is used to solve the multi objective airfoil optimization problem. And the satisfying Pareto front is obtained.

Key words: genetic algorithm, Pareto optimum, multi-objective optimization, airfoil