王威, 王军, 梁钟, 李佳峻. 基于熵产理论设计方法的多目标翼型优化[J]. 空气动力学学报, 2019, 37(5): 827-833. DOI: 10.7638/kqdlxxb-2017.0055
引用本文: 王威, 王军, 梁钟, 李佳峻. 基于熵产理论设计方法的多目标翼型优化[J]. 空气动力学学报, 2019, 37(5): 827-833. DOI: 10.7638/kqdlxxb-2017.0055
WANG Wei, WANG Jun, LIANG Zhong, LI Jiajun. Entropy generation theory based design methodology for multi-objective airfoil shape optimization[J]. ACTA AERODYNAMICA SINICA, 2019, 37(5): 827-833. DOI: 10.7638/kqdlxxb-2017.0055
Citation: WANG Wei, WANG Jun, LIANG Zhong, LI Jiajun. Entropy generation theory based design methodology for multi-objective airfoil shape optimization[J]. ACTA AERODYNAMICA SINICA, 2019, 37(5): 827-833. DOI: 10.7638/kqdlxxb-2017.0055

基于熵产理论设计方法的多目标翼型优化

Entropy generation theory based design methodology for multi-objective airfoil shape optimization

  • 摘要: 将熵产理论引入翼型的多目标气动优化设计,通过理论推导湍流流场熵产率计算公式,进而阐述熵分析方法在翼型气动优化中的作用。通过CST参数化方法对翼型进行参数化建模,并将多目标优化算法与CFD计算耦合起来,计算翼型升阻比和流场熵产率,建立一种以最大升阻比和最小熵产率为目标的翼型优化方法,进而得到优化翼型的Pareto解,并与传统翼型优化方法进行对比。与初始种群相比,优化翼型具有更优的气动性能,在升阻比提高的条件下,流场熵产率减少,能量效率提高。采用多目标遗传算法得到的非支配解集分布均匀,质量较高,设计者可以根据设计需要选择具有相对低熵产的一组翼型来改善空气动力性能。结果分析表明,本文所建方法具有较强的全局收敛性,具有一定的工程应用前景。

     

    Abstract: In this paper, an entropy analysis and design optimization methodology is combined with multi-objective airfoil shape optimization to demonstrate the impact of entropy generation on aerodynamic designs. Model equations for the calculation of the local entropy generation rate in turbulent flows are presented by extending the Reynolds-averaging procedure to the entropy balance equation. We use the CST parametric method to model the airfoil configuration and combine the CFD solver with multi-objective optimization algorithm respectively, to compute aerodynamics quantities such as lift-drag ratio and entropy generation rate. We obtain the optimized airfoils that maximize lift-drag ratios and minimize entropy generation rates as optimization objectives. Moreover, the present method is compared with the traditional method. From the multi-objective solutions, a designer can select a set of airfoil shapes with low relative entropy generation, small actuation cost, and improved aerodynamic performance at the design conditions. The optimized results unfold that, the proposed method is capable of realizing better performance in practical aerodynamic optimization, with the features of strong global convergence and improved aerodynamic performance in the increase of lift-drag ratio and the decrease of entropy generation rate, compared with the initial population.

     

/

返回文章
返回