航空学报 > 2019, Vol. 40 Issue (2): 522381-522381   doi: 10.7527/S1000-6893.2018.22381

动力干扰下宽体客机机翼多目标优化设计

薛帮猛, 张文升, 孙学卫, 吴宇昂   

  1. 中国商飞北京民用飞机技术研究中心 民用飞机设计数字仿真技术北京市重点实验室, 北京 102211
  • 出版日期:2019-02-15 发布日期:2018-07-13
  • 通讯作者: 薛帮猛 E-mail:xuebangmeng@comac.cc
  • 基金资助:
    国家级项目

Multi-objective wing shape optimization for a wide-body civil aircraft in wing-body-pylon-powered nacelle configuration

XUE Bangmeng, ZHANG Wensheng, SUN Xuewei, WU Yuang   

  1. Beijing Key Laboratory of Simulation Technology for Civil Aircraft Design, COMAC Beijing Aeronautical Science and Technology Research Institute, Beijing 102211, China
  • Online:2019-02-15 Published:2018-07-13
  • Supported by:
    National Level Project

摘要: 在机翼/机身/吊挂/动力短舱(WBPN)构型中开展了宽体客机机翼外形多目标优化设计。通过对动力短舱流场的动量积分,分析了直接用壁面积分"阻力"作为机体外形减阻优化设计目标函数的合理性。计算研究了短舱/吊挂的安装,以及发动机喷流对翼吊布局宽体客机机翼的干扰作用,展示了同时在安装效应和喷流干扰下设计机翼外形的重要性。运行搭建于超级计算机上的优化系统,求解雷诺平均Navier-Stokes(RANS)方程计算流场,实现了动力干扰下机翼外形的三点三目标优化设计。在80 h内,完成了近20 000个方案的计算评估,遗传优化近40代。所选的最优方案阻力发散性能明显提高,自动优化后的人工修形设计使机翼剖面展向过渡和压力分布形态更为理想。动力构型下取得的减阻效果,在通气短舱构型下亦得到验证和确认。

关键词: 机翼/机身/吊挂/动力短舱构型, 多目标优化设计, 喷流干扰, 雷诺平均Navier-Stokes(RANS)方程, 超级计算机

Abstract: A recent effort of multi-objective wing shape optimization for a wide-body civil aircraft in Wing-Body-Pylon-powered Nacelle (WBPN) configuration is presented. Based on the momentum analysis for the flow-field of a powered-on configuration, the nominal "drag" directly from wall integration can be reasonably used as the cost function in airframe shape optimizations without thrust/drag bookkeeping. The lift losses of the baseline shape induced by pylon/nacelle and jet are calculated, and the results show the importance of considering both the installation and jet effects when designing wing shape of a wing-mounted wide-body aircraft. By running an automated optimization framework setup on a super computer, a three objective wing shape optimization process for the powered-on configuration is completed in 80 h. Nearly twenty thousand cases with 40 generations of evolution are evaluated by solving Reynolds Averaged Navier-Stokes (RANS) equations. The selected optimum has better performance of drag divergence than the baseline. A manual refinement process after automatic optimization improves both spanwise thickness variation and the distribution of sectional pressure. The drag reduction achieved under power-on configuration is also affirmed by the verification calculation under the flow through configuration.

Key words: wing-body-pylon-powered nacelle configuration, multi-objective optimization design, jet interference, Reynolds Averaged Navier-Stokes (RANS) equations, super computers

中图分类号: