航空学报 > 2023, Vol. 44 Issue (5): 126434-126434   doi: 10.7527/S10006893.2021.26434

基于卷积神经网络的结冰翼型气动特性建模

何磊1,2, 钱炜祺1,2, 董康生2, 易贤1(), 柴聪聪1   

  1. 1.中国空气动力研究与发展中心 空气动力学国家重点实验室,绵阳  621000
    2.中国空气动力研究与发展中心 计算空气动力研究所,绵阳  621000
  • 收稿日期:2021-09-24 修回日期:2021-10-15 接受日期:2021-11-08 出版日期:2023-03-15 发布日期:2021-11-12
  • 通讯作者: 易贤 E-mail:yixian_2000@163.com
  • 基金资助:
    国家级项目

Aerodynamic characteristics modeling of iced airfoil based on convolution neural networks

Lei HE1,2, Weiqi QIAN1,2, Kangsheng DONG2, Xian YI1(), Congcong CHAI1   

  1. 1.State Key Laboratory of Aerodynamics,China Aerodynamics Research and Development Center,Mianyang  621000,China
    2.Computational Aerodynamics Research Institute,China Aerodynamics Research and Development Center,Mianyang  621000,China
  • Received:2021-09-24 Revised:2021-10-15 Accepted:2021-11-08 Online:2023-03-15 Published:2021-11-12
  • Contact: Xian YI E-mail:yixian_2000@163.com
  • Supported by:
    National Level Project

摘要:

提出了基于卷积神经网络(CNN)的结冰翼型气动特性预测方法,设计了输入层结冰翼型图像规范,克服了复杂冰形在翼面同一位置法线方向存在多值,单值函数难以描述的问题。预测模型可同时预测多个迎角对应的升阻力系数,实现了直接从冰形图像到气动特性的快速预测,对升力系数和阻力系数预测结果的平均相对误差均可控制在8%以内。重点研究了不同卷积层数量、卷积核数量、卷积核尺寸对模型性能的影响规律:CNN的不同层次特征对应不同滤波频率,卷积层数增加会捕获更多高频特征量;增加卷积核数量可提取更多冰形特征,提升模型性能,但数量过多会增加冗余特征,降低模型泛化性能;阻力系数预测模型对卷积核数量的最低要求大于升力系数,其原因在于,相较升力系数,阻力系数不仅受翼面压差影响,还受摩阻特性影响,其建模所需的关键特征数量多于升力系数;增大卷积核尺寸,可扩大卷积操作“视野”,增强对冰形整体特征信息的提取,有利于提升模型泛化性能。相关结论为飞机结冰气动特性实时动态预测与监测提供了新的思路和方法支撑。

关键词: 飞机结冰, 气动特性, 机器学习, 深度学习, 卷积神经网络

Abstract:

A prediction method for iced airfoil aerodynamic characteristics is proposed based on the Convolutional Neural Networks (CNN). This new method is able to solve the multiple-value problem of the complex ice shape along the wall-normal direction at a specific location of the airfoils. The CNN prediction model can directly predict the lift and drag coefficient at multiple angles of attacks via a rapid process from the iced airfoil image to aerodynamics characteristics. The relative mean errors of the resulting lift coefficient and drag coefficient are smaller than 8%. By comparison, the influence of the convolutional layer number, filter number, and filter size on the model performance is investigated. The features of different layers in the CNN correspond to different filter frequencies, and increasing the number of convolutional layers will capture more features of high frequency. With the increase in the number of filters, more features of the ice shape will be extracted, promoting the model performance. However, more redundant features will appear if the number of filters exceeds a critical value, resulting in degradation of the generalization performance. The required number of filters for drag coefficient prediction is larger than that for lift coefficient, because drag coefficient is affected by both the pressure difference on the airfoil and friction, leading to a larger number of key features required for drag coefficient prediction than that for lift coefficient. Moreover, a larger size of filters will widen the field of view of convolution operation, enhancing the capability of extracting global features, and promoting the generalization performance of the model. The current new method can be applied to real-time prediction and monitor of aerodynamic characteristics of aircraft icing.

Key words: aircraft icing, aerodynamic characteristics, machine learning, deep learning, convolutional neural networks

中图分类号: