航空学报 > 2025, Vol. 46 Issue (20): 531920-531920   doi: 10.7527/S1000-6893.2025.31920

基于声爆高效预测的超声速民机优化设计与数据挖掘

李祚泰1,2, 陈树生1,2(), 金世轶1,2, 高正红1,2, 周卫国2,3   

  1. 1.西北工业大学 航空学院,西安 710072
    2.飞行器基础布局全国重点实验室,西安 710072
    3.航空工业第一飞机设计研究院,西安 710089
  • 收稿日期:2025-03-03 修回日期:2025-04-01 接受日期:2025-05-18 出版日期:2025-05-28 发布日期:2025-05-27
  • 通讯作者: 陈树生 E-mail:sshengchen@nwpu.edu.cn

Optimization design and data mining for supersonic civil aircraft based on sonic boom efficient prediction

Zuotai LI1,2, Shusheng CHEN1,2(), Shiyi JIN1,2, Zhenghong GAO1,2, Weiguo ZHOU2,3   

  1. 1.School of Aeronautics,Northwestern Polytechnical University,Xi’an 710072,China
    2.National Key Laboratory of Aircraft Configuration Design,Xi’an 710072,China
    3.AVIC The First Aircraft Design Institute,Xi’an 710089,China
  • Received:2025-03-03 Revised:2025-04-01 Accepted:2025-05-18 Online:2025-05-28 Published:2025-05-27
  • Contact: Shusheng CHEN E-mail:sshengchen@nwpu.edu.cn

摘要:

绿色超声速民机是未来民用飞行器的重要发展方向,声爆是制约其发展的关键因素之一。传统的CFD方法计算量大、耗时长,难以直接用于初期迭代设计;基于面积律的声爆高效预测方法能够在保证一定精度的前提下大幅缩短计算时间,契合了超声速民机初期设计阶段大量、快速的计算需求。因此,首先基于声爆面积律高效预测方法,建立了一套超声速民机低声爆优化设计框架。然后以LM1021无短舱构型为初始构型,验证了多声爆传播角下优化设计的有效性,优化结果均经过高精度CFD验证:0°传播角优化下的远场最大过压降低了32.03%,0°和30°传播角协同优化下的远场最大过压分别降低了30.45%和32.66%。此外借助声爆高效预测方法快速生成超声速民机声爆数据集,采用随机森林和自适应增强算法进行智能数据挖掘,提取超声速民机低声爆关键设计变量。最后经对比验证,数据挖掘结果与CFD流场特征、面积律理论相吻合。结果表明,合理设计机鼻静音锥、翼根翼型及上反角、机尾外形是超声速民机低声爆设计的关键。基于高效预测方法开展的声爆优化设计和数据挖掘表现出显著的效率优势,能够为后续的超声速民机设计工作提供技术支撑。

关键词: 超声速民机, 声爆预测, 面积律, 优化设计, 数据挖掘

Abstract:

Green SuperSonic Transport (SST) is a key future direction in civil aircraft, with sonic boom remaining an important factor affecting its development. Traditional CFD method is unsuitable for massive and iterative designs, due to large computation quantity and time-consumption. Sonic boom efficient prediction method based on area rule can significantly reduce computation time with a certain level of accuracy, making it suitable for the requirement of a large number of rapid calculations in the initial design phase of SST. Therefore, based on the efficient prediction method, we first establishe a framework for low sonic boom optimization design of SST. Then LM1021 no-nacelle configuration is optimized to verify the optimization in multiple propagation angles. The optimization results are validated by high-accuracy CFD method: the maximum overpressure in the far field is reduced by 32.03% considering the propagation angle of 0°, while the maximum overpressures are decreased by 30.45% and 32.66% respectively when it takes the angles of 0° and 30° into account together. In addition,sonic boom dataset of SST is generated with the help of efficient prediction method. Random forest and AdaBoost algorithms are applied for intelligent data mining to extract the key design variables of low sonic boom. Finally, the results of data mining are consistent with CFD flow fields and area rule, highlighting that the designs of the Quiet Spike, the airfoil and dihedral angle of wing root, and the tail shape are crucial for sonic boom mitigation. The combination of sonic boom optimization and data mining based on efficient prediction method demonstrates advantage of high efficiency, offering technical support for future SST design.

Key words: supersonic civil aircraft, sonic boom prediction, area rule, optimization design, data mining

中图分类号: