航空学报 > 2025, Vol. 46 Issue (8): 630778-630778   doi: 10.7527/S1000-6893.2024.30778

基于快速逐线计算模型的高超声速羽流红外辐射计算方法

徐建宇1, 周莉1,2(), 王占学1,2, 是介1, 史毫1   

  1. 1.西北工业大学 动力与能源学院,西安 710129
    2.先进航空发动机协同创新中心,北京 100191
  • 收稿日期:2024-06-03 修回日期:2024-06-27 接受日期:2024-08-12 出版日期:2024-08-26 发布日期:2024-08-20
  • 通讯作者: 周莉 E-mail:zhouli@nwpu.edu.cn
  • 基金资助:
    国家自然科学基金(52376032);陕西省杰出青年科学基金(2021JC-10);国家科技重大专项(J2019-Ⅱ-0015-0036);航空发动机及燃气轮机基础科学中心项目(P2022-B-I-002-001,P2022-B-Ⅱ-010-001);中央高校基本科研业务费专项资金(501XTCX2023146001)

Calculation method for hypersonic plume infrared radiation based on a fast line-by-line calculation model

Jianyu XU1, Li ZHOU1,2(), Zhanxue WANG1,2, Jie SHI1, Hao SHI1   

  1. 1.School of Power and Energy,Northwestern Polytechnical University,Xi’an 710129,China
    2.Collaborative Innovation Center for Advanced Aero-Engine,Beijing 100191,China
  • Received:2024-06-03 Revised:2024-06-27 Accepted:2024-08-12 Online:2024-08-26 Published:2024-08-20
  • Contact: Li ZHOU E-mail:zhouli@nwpu.edu.cn
  • Supported by:
    National Natural Science Foundation of China(52376032);Funds for Distinguished Young Scholars of Shaanxi Province(2021JC-10);National Science and Technology Major Project (J2019-?Ⅱ?-0015-0036);Science Center for Gas Turbine Project (P2022-B-I-002-001, P2022-B-?Ⅱ?-010-001);the Fundamental Research Funds for the Central Universities(501XTCX2023146001)

摘要:

为对临近空间高超声速飞行器羽流的红外辐射特性进行研究,构建了适用于临近空间羽流流场的红外辐射特性计算方法。首先采用非线性耦合本构关系模型对羽流流场进行数值仿真,检验了流场计算模型的准确性。其次采用全因子试验设计,通过逐线计算物理模型获取了46 200组流场特性参数与吸收系数的数据,训练获得基于BP神经网络的快速逐线计算模型,模型最大平均绝对误差和决定系数分别为0.003 65和0.999 4,相比逐线计算物理模型计算速度提升了4个数量级。最后结合逆蒙特卡洛方法作为红外辐射传输方法,建立了高超声速飞行器羽流的红外辐射计算方法。采用建立的流场和红外辐射计算方法针对类X-51A飞行器巡航状态下羽流红外辐射特性进行了研究。研究表明,临近空间下高超声速羽流呈现出外热流包裹冷内流的现象,羽流边缘滞止区红外辐射更强,红外成像中呈现“剪刀”状,剪切层的变化影响羽流红外成像的特征。

关键词: 高超声速羽流, 非线性耦合本构关系, 逐线计算法, 深度学习, 逆向蒙特卡洛方法, 红外辐射

Abstract:

To investigate the infrared radiation characteristics of plumes of hypersonic vehicles in near space, a computational methodology for the flow field of the near space plume infrared radiation was developed. Initially, the numerical simulation of the plume flow field is carried out by using the nonlinear coupling constitutive relationship model, and the accuracy of the flow field calculation model is verified. Subsequently, a comprehensive dataset consisting of 46 200 sets of flow field characteristic parameters and absorption coefficients was generated using a line-by-line calculation physical model, informed by a full-factorial experimental design. This dataset was employed to train an accelerated line-by-line calculation model based on a BP neural network. The trained model demonstrated a maximum mean absolute error of 0.003 65 and an R2 value of 0.999 4, achieving computational speeds four times faster than the traditional line-by-line calculation physical model. Finally, combined with the Backward Monte Carlo method as the infrared radiation transmission method, the infrared radiation calculation method of the hypersonic vehicles plume flow field is established. The infrared radiation characteristics of the plume under the cruising state of X-51A aircraft were studied by using the established flow field and infrared radiation calculation methods. The findings indicate that hypersonic plumes in near space exhibit a phenomenon where external heat flow envelops a cooler internal flow. Furthermore, the infrared radiation is stronger in the stagnation region at the plume’s periphery, resulting in a “scissors” pattern in infrared imaging. Variations in the shear layer significantly influence the infrared imaging characteristics of the plume.

Key words: hypersonic plume, nonlinear coupled constitutive relations, line by line method, deep learning, backward Monte Carlo method, infrared radiation

中图分类号: