Differential Evolution Algorithm Study And Application In Aerodynamic Optimization Design

Aerodynamic optimization based on CFD has been an important field in current CFD research. At present, many optimization methods have been successfully applied in aerodynamic design of airfoil and wing, with very good effect. In order to apply aerodynamic optimization design better in engineer practice, optimization method research is a key facet, in addition to improvement of the precision and efficiency for the solving N-S equation. The inbeing of traditional determinate optimization method determines the local optimization character of optimization result, while evolutional algorithm has an advantage of global optimization, besides its high robust and parallelity, which causes its more and more extensive application in engineering optimization. In recent years, a new evolutional algorithm ---differential evolutional algorithm has caught comprehensive attention of scholars of all countries, with its main character of briefness, quick convergence, and few knowledge of related problem. A mass of research shows, differential evolutional algorithm has good convergence ability, and is very suitable for solving complicated optimization problems. This thesis studies on differential evolutional algorithm and makes some modification. Then some difference evolutional optimization models for aerodynamic optimization design were established, and were applied in aerodynamic optimization design of airfoil and wing. The main work of the thesis is processed on the following facets:1. All-sided introduction of the principle, basic structure, algorithm character and expanding mode etc.; In-depth study on the choosing of parameter of differential evolutional algorithm is processed, in testing of standard function.2. Fulfillment of aerodynamic optimization design of airfoil and wing, in combination of flow solving program based on 2-dimensional and 3-dimensional equation and differential evolutional algorithm. Getting exact computational result, in comparison of traditional engineering method and optimization design based on potential theory.3. Forming Pareto differential evolutional algorithm which can efficiently solve multi-objective optimization design problem, in combination of Pareto method and differential evolutional algorithm. In aerofoil optimization design, Pareto differential evolutional algorithm is used for multi-object optimization design.4. Great improvement of computational efficiency, in parallelization of differential evolutional algorithm in MPI method, which demonstrates its good ability.Some helpful conclusion, which is drawn of practical study on aerodynamic optimization design in use of differential evolutional algorithm, can be referenced for airplane aerodynamic optimization design based on numerical computational method.