| Popis: |
This paper presents a novel approach of two evolutionary algorithms to build a model that describes the geometric dynamics of aerodynamic airfoils by fitting a parametric curve. The proposal allows determining the best parameters of a polynomial that fits the geometry of an airfoil by analyzing the space of solutions by a Monte Carlo type sampling model. The composition of genetic Algorithms allows to determine the optimal values for the polynomial fit of the model in two stages: 1) the selection of the best reference points to model a polynomial that represents the geometry, and 2) the determination of the degree of the polynomial that best fits the geometry of aerodynamic airfoil discarding the local or sub-optimal solutions. The proposal becomes validated with profiles of the blade, a model for an offshore wind turbine in several scenarios, showing where the dynamic description of the airfoils determines the difficulty of an optimal fitting. Finally, this approach allows the construction of airfoil models with a resolution of 0.005224 error in the adjustment for the generation of the airfoil model. Keywords: Genetic algorithm, Aerodynamic airfoil, Optimal fitting, Automatic model generation |
