A fast and efficient optimization method is proposed for winglet design optimization. The optimization method starts from a random population, whose variables are normalized in the range $[0,1]$. Population members are then divided into two expert groups: the free group and the guided group; each has specific tasks for the active search of the domain but with a single operator. The deliberate search of the expert groups and proper adjustment of diversity and intensification leads to rapid population direction towards global optimization. The global performance of the method is first evaluated in two standard problems. Comparison of the results with alternative methods shows that the proposed method is superior in terms of convergence speed and accuracy. Then, a winglet design optimization problem for the DLR-F6 wing is carried out. The objective function evaluation is done by a high-fidelity computational fluid dynamics method. A robust CAD-based method is used for winglet shape parameterization. Results show that the new method is able to outperform particle swarm optimization (PSO), genetic algorithm (GA), and Mean-Variance Mapping Optimization (MVMO). The optimal winglet achieves a drag coefficient reduction of 9.19% compared to the initial wing while keeping the lift coefficient unchanged. Then, the effect of winglet design variables on the optimization process is investigated. It is found that the best result with minimum computational cost is achieved when using only three winglet shape parameters of length, twist angle and angle of attack. Finally, the root bending moment minimization is added to the objective function as a structural consideration, and its effects on the results are studied.

提出了一种快速有效的小翼设计优化方法。优化方法从随机总体开始，其变量在范围内标准化$[0,1]$. 然后将人口成员分为两个专家组：自由组和指导组；每个都有用于主动搜索域的特定任务，但只有一个操作员。专家组的深思熟虑和多样性和集约化的适当调整导致种群快速走向全局优化。该方法的全局性能首先在两个标准问题中进行评估。结果与替代方法的比较表明，所提出的方法在收敛速度和精度方面具有优越性。然后，进行了DLR-F6机翼的小翼设计优化问题。目标函数评估是通过高保真计算流体动力学方法完成的。一种稳健的基于 CAD 的方法用于小翼形状参数化。结果表明，新方法的性能优于粒子群优化（PSO）、遗传算法（GA）和均值方差映射优化（MVMO）。与初始机翼相比，最佳小翼实现了 9.19% 的阻力系数降低，同时保持升力系数不变。然后，研究了小翼设计变量对优化过程的影响。结果表明，当仅使用长度、扭转角和攻角这三个小翼形状参数时，可以以最小的计算成本获得最佳结果。最后，将根弯矩最小化作为结构考虑添加到目标函数中，并研究其对结果的影响。与初始机翼相比，最佳小翼实现了 9.19% 的阻力系数降低，同时保持升力系数不变。然后，研究了小翼设计变量对优化过程的影响。结果表明，当仅使用长度、扭转角和攻角这三个小翼形状参数时，可以以最小的计算成本获得最佳结果。最后，将根弯矩最小化作为结构考虑添加到目标函数中，并研究其对结果的影响。与初始机翼相比，最佳小翼实现了 9.19% 的阻力系数降低，同时保持升力系数不变。然后，研究了小翼设计变量对优化过程的影响。结果表明，当仅使用长度、扭转角和攻角这三个小翼形状参数时，可以以最小的计算成本获得最佳结果。最后，将根弯矩最小化作为结构考虑添加到目标函数中，并研究其对结果的影响。结果表明，当仅使用长度、扭转角和攻角这三个小翼形状参数时，可以以最小的计算成本获得最佳结果。最后，将根弯矩最小化作为结构考虑添加到目标函数中，并研究其对结果的影响。结果表明，当仅使用长度、扭转角和攻角这三个小翼形状参数时，可以以最小的计算成本获得最佳结果。最后，将根弯矩最小化作为结构考虑添加到目标函数中，并研究其对结果的影响。