亚声速无人机S弯进气道的多点多目标优化设计

Multi-Points and Multi-Objective Optimization Design of Subsonic UAV S-Shaped Inlet

关键词:  S弯进气道  多点优化设计  总压恢复  总压畸变  数值模拟

Key words:  S-shaped inlet  Multi-points optimization design  Total pressure recovery  Total pressure distortion  Numerical simulation

摘要:

为提高亚声速无隔道式S弯进气道的整体气动性能，以提高总压恢复系数和降低畸变指数为设计目标，结合高精度数值模拟方法与第二代非劣排序遗传算法（NSGA-II），开展了无隔道式S弯进气道在马赫数0.25和0.7时的多目标优化设计。整个优化流程基于400个样本，最终得到四幅有效Pareto前沿图。从总压畸变Pareto前沿图中选取出优化算例并与原始进气道进行对比，结果表明：优化后的进气道中心线斜率入口段小、出口段大，而横截面面积分布的曲线斜率恰好相反；优化后的进气道低压区缩小、流动分离得到有效控制；虽然总压恢复系数提高有限，但是总压畸变得到大幅降低，在马赫数为0.25和0.7时，分别降低15.86%和23.61%。优化后的进气道在马赫数0.25~0.7整体性能得到有效改善。将该优化设计方法进一步推广应用于3个马赫数下的多点多目标优化设计，并得到了三维Pareto前沿图。

Abstract:

In order to increase the performance of subsonic diverterless S-shaped inlet within a range of Mach number， a multi-objective optimization system for S-shaped diverterless inlet was established， and the Mach numbers are 0.25 and 0.7. With the design objective of increasing total pressure recovery coefficient and decreasing distortion， the optimization system combined the numerical simulation method with Non-dominated Sorting Genetic Algorithm II （NSGA-II）. The optimization case was based on 400 samples and four Pareto Front figures were obtained. An optimized design was chosen from the distortion Pareto Front figure. Compared with the original inlet， the slope of the optimized centerline is smaller at the entrance and larger at the exit， while the distribution of the optimized cross section area is reverse. Low pressure areas are obviously shrunk and flow separation is effectively reduced after optimization. Although the improvement of the total pressure recovery coefficient is limited， the total pressure distortion is significantly redued， with a decrease of 15.86% at Ma 0.25 and 23.61% at Ma 0.7， respectively. In conclusion， the overall performance of the optimized inlet is effectively improved within Mach number 0.25~0.7. The optimization design method has been applied to multi-objective optimization design under three Mach numbers， and several 3D Pareto Front figures are obtained.