遗传算法结合神经网络的多段翼型优化设计研究

为提高多段翼型增升效能，开展包括襟／缝翼偏度和缝道参数在内的优化设计研究。将神经网络与遗传算法结合的优化设计方法应用于气动优化设计，并针对30P30N三段翼型，分别以8°迎角时升阻比最大和22°迎角时升力最大为目标进行了单目标和多目标优化设计研究。研究结果表明：采用单目标设计虽可在设计点获得较好的优化结果，但在非设计状态气动性能下降；采用多目标优化设计，既可获得良好的中等迎角升阻性能，又可改善大迎角失速性能，使综合气动性能更优；遗传算法与神经网络结合的优化设计方法可满足多段翼型的多点优化设计问题，具有高效、高精度等优点，易于工程应用。

Optimization Design for Multi-element Airfoils Based on Genetic Algorithms and Neural Networks

To improve the aerodynamic performance of multi-element airfoils, the optimization of parameters in cluding the deflection angle, gap and overlap of flaps and leading edge slats are conducted. The optimization works are based on the combination of genetic algorithm and artificial neural network. Single objective and multi-objective optimization design for 30P30N airfoil are carried out respectively so as to improve lift to-drag ratio at angle 8°and lift at angle 22°. Single objective optimization shows that, the aerodynamic characteristics at design points are improved but it can hardly satisfy the requirements at off-design points. So, a multi-objective optimization design is done to improve lift-to-drag at medium angle of attack and lift at high angle of attack. The investigation demonstrates that genetic algorithm and artificial neural network can satisfy multi-objective optimi zation design of multi element airfoils with high efficiency and precision, and moreover, this method can be easi ly applied in engineering.