留言板

尊敬的读者、作者、审稿人, 关于本刊的投稿、审稿、编辑和出版的任何问题, 您可以本页添加留言。我们将尽快给您答复。谢谢您的支持!

姓名
邮箱
手机号码
标题
留言内容
验证码

基于机器学习的结冰风洞温度场预测

张兴焕 张平涛 彭博 易贤

张兴焕,张平涛,彭博,等. 基于机器学习的结冰风洞温度场预测[J]. 实验流体力学,2022,36(5):8-15 doi: 10.11729/syltlx20210196
引用本文: 张兴焕,张平涛,彭博,等. 基于机器学习的结冰风洞温度场预测[J]. 实验流体力学,2022,36(5):8-15 doi: 10.11729/syltlx20210196
ZHANG X H,ZHANG P T,PENG B,et al. Prediction of icing wind tunnel temperature field with machine learning[J]. Journal of Experiments in Fluid Mechanics, 2022,36(5):8-15. doi: 10.11729/syltlx20210196
Citation: ZHANG X H,ZHANG P T,PENG B,et al. Prediction of icing wind tunnel temperature field with machine learning[J]. Journal of Experiments in Fluid Mechanics, 2022,36(5):8-15. doi: 10.11729/syltlx20210196

基于机器学习的结冰风洞温度场预测

doi: 10.11729/syltlx20210196
基金项目: 国家重大科技专项(2019-III-0010-0054)
详细信息
    作者简介:

    张兴焕:(1998—),女,四川达州人,硕士。研究方向:机器学习,深度学习,结冰风洞温度场预测。通信地址:四川省成都市新都区新都大道8号西南石油大学(610500)E-mail:1183327823@qq.com

    通讯作者:

    E-mail:bopeng@swpu.edu.cn

    yixian_2000@163.com

  • 中图分类号: V211.24

Prediction of icing wind tunnel temperature field with machine learning

  • 摘要: 结冰风洞是开展飞行器结冰与防除冰研究的重要基础设施,其制冷系统通过调节压缩机吸气压力实现风洞内气流温度的精确控制,吸气压力控制及降温方式影响着风洞的试验效率。为实现压缩机吸气压力的准确预测,本文采用自适应粒子群算法优化后的支持向量回归(APSO–SVR)建立预测模型;在此基础上,利用多层感知机(MLP)神经网络建立分析模型,研究试验工况参数对风洞降温速率的影响。结果表明:压缩机吸气压力的预测值与试验值的平均绝对百分比误差(EMAP)低于4%,均方误差(EMS)低于0.003;影响风洞降温速率的工况参数主要有气流压力、试验风速、压缩机吸气压力和换热器出口初始温度,其中,压缩机吸气压力对降温速率的影响是最显著的。
  • 图  1  3 m×2 m结冰风洞[11]

    Figure  1.  The CARDC icing tunnel

    图  2  制冷系统管路流程图

    Figure  2.  Refrigeration system pipeline flow chart

    图  3  温度控制结构图

    Figure  3.  Temperature control structure diagram

    图  4  APSO算法流程图

    Figure  4.  Flow chart of APSO algorithm

    图  5  SVR算法示意图

    Figure  5.  The schematic diagram of SVR

    图  6  APSO-SVR模型预测值与试验值对比

    Figure  6.  The comparison between predicted result of APSO-SVR model and experimental result

    图  7  MLP网络结构图

    Figure  7.  The network structure of MLP

    图  8  MLP模型预测值与试验值对比

    Figure  8.  The comparison between predicted result of MLP model and experimental result

    图  9  降温速率变化曲线(p=191 kPa)

    Figure  9.  Change curve of cooling rate when pressure is 191 kPa

    图  10  降温速率变化曲线(p=202 kPa)

    Figure  10.  Change curve of cooling rate when pressure is 202 kPa

    图  11  2个工况的下降温速率变化趋势曲线

    Figure  11.  Change trend curve of cooling rate under 2 working conditions

    表  1  压缩机吸气压力试验值和预测值

    Table  1.   Compressor suction pressure test value and prediction result

    工况v/(m·s−1p1/kpaTin/℃Tout/℃p吸,试/kPap吸,预/kPaEAP/%
    1 140.88 69.34 −6.21 −12.4 89 87 2.51
    2 78.85 91.76 −1.73 −4.91 193 195 0.94
    3 139.79 46.04 −0.26 −5.28 177 174 1.89
    4 39.09 94.27 −2.32 −5.18 193 196 1.51
    5 77.03 94.55 −1.18 −4.97 185 184 0.32
    6 99.33 90.30 −0.78 −4.04 194 198 2.04
    下载: 导出CSV

    表  2  消融试验结果

    Table  2.   The result of ablation experiment

    模型EMAP
    MLP15.1%
    MLP−v35.7%
    MLP−p117.4%
    MLP−p65.1%
    MLP−T026.8%
    下载: 导出CSV

    表  3  工况数据表

    Table  3.   Working condition data sheet

    工况v/(m·s−1T0/℃p1/kpaTin/℃t/s
    160.50−0.5125046.370.0350476
    2140.15−8.2812569.64−1.9375242
    下载: 导出CSV
  • [1] 林贵平, 卜雪琴, 申晓斌. 飞机结冰与防冰技术[M]. 北京: 北京航空航天大学出版社, 2016.

    LIN G P, BU X Q, SHEN X B. Aircraft icing and anti-icing technology[M]. Beijing: Beijing University of Aeronautics & Astronautics Press, 2016.
    [2] VAN-ZANTE J, IDE R, STEEN L C. NASA Glenn icing research tunnel: 2012 cloud calibration procedure and results[C]//Proc of the 4th AIAA Atmospheric and Space Environments Conference. 2012: 2933. doi: 10.2514/6.2012-2933 http://dx. doi.org/10.2514/6.2012-2933
    [3] 郭向东,张平涛,张珂,等. 3 m × 2 m结冰风洞热流场品质提高及评估[J]. 实验流体力学,2021,35(4):41-51. doi: 10.11729/syltlx20200118

    GUO X D,ZHANG P T,ZHANG K,et al. Improvement and evaluation of thermal flow-field quality in CARDC icing wind tunnel[J]. Journal of Experiments in Fluid Mechanics,2021,35(4):41-51. doi: 10.11729/syltlx20200118
    [4] 柴聪聪,王强,易贤,等. 基于卷积神经网络的结冰翼型气动参数预测[J]. 飞行力学,2021,39(5):13-18. doi: 10.13645/j.cnki.f.d.20210811.001

    CHAI C C,WANG Q,YI X,et al. Aerodynamic parameters prediction of airfoil ice accretion based on convolutional neural network[J]. Flight Dynamics,2021,39(5):13-18. doi: 10.13645/j.cnki.f.d.20210811.001
    [5] 陈海,钱炜祺,何磊. 基于深度学习的翼型气动系数预测[J]. 空气动力学学报,2018,36(2):294-299.

    CHEN H,QIAN W Q,HE L. Aerodynamic coefficient prediction of airfoils based on deep learning[J]. Acta Aerodynamica Sinica,2018,36(2):294-299.
    [6] 何磊,钱炜祺,易贤,等. 基于转置卷积神经网络的翼型结冰冰形图像化预测方法[J]. 国防科技大学学报,2021,43(3):98-106. doi: 10.11887/j.cn.202103013

    HE L,QIAN W Q,YI X,et al. Graphical prediction method of airfoil ice shape based on transposed convolution neural networks[J]. Journal of National University of Defense Technology,2021,43(3):98-106. doi: 10.11887/j.cn.202103013
    [7] OGRETIM E,HUEBSCH W,SHINN A. Aircraft ice accretion prediction based on neural networks[J]. Journal of Aircraft,2006,43(1):233-240. doi: 10.2514/1.16241
    [8] CHANG S N,LENG M Y,WU H W,et al. Aircraft ice accretion prediction using neural network and wavelet packet transform[J]. Aircraft Engineering and Aerospace Technology,2016,88(1):128-136. doi: 10.1108/aeat-05-2014-0057
    [9] SURESH S,OMKAR S N,MANI V,et al. Lift coefficient prediction at high angle of attack using recurrent neural network[J]. Aerospace Science and Technology,2003,7(8):595-602. doi: 10.1016/S1270-9638(03)00053-1
    [10] 兰其龙. 基于神经网络预测控制的低温风洞多变量控制策略研究[D]. 绵阳: 中国空气动力研究与发展中心, 2016.

    LAN Q L. Research on multivariable control strategy based on neural network predictive control in cryogenic wind tunnel[D]. Mianyang: China Aerodynamics Research and Development Center, 2016.
    [11] 郭向东,柳庆林,赖庆仁,等. 大型结冰风洞气流场适航符合性验证[J]. 空气动力学学报,2021,39(2):184-195. doi: 10.7638/kqdlxxb-2019.0086

    GUO X D,LIU Q L,LAI Q R,et al. Airworthiness compliance verification of aerodynamic flowfield of a large-scale icing wind tunnel[J]. Acta Aerodynamica Sinica,2021,39(2):184-195. doi: 10.7638/kqdlxxb-2019.0086
    [12] 郭向东,张平涛,赵献礼,等. 大型结冰风洞热流场符合性验证[J]. 实验流体力学,2020,34(5):79-88. doi: 10.11729/syltlx20190113

    GUO X D,ZHANG P T,ZHAO X L,et al. The compliance verification of thermodynamic flowfield in the large icing wind tunnel[J]. Journal of Experiments in Fluid Mechanics,2020,34(5):79-88. doi: 10.11729/syltlx20190113
    [13] KENNEDY J, EBERHART R C. Particle swarm optimization[C]//Proc of the Proceedings of ICNN'95 - International Conference on Neural Networks. 1995. doi: 10.1109/ICNN. 1995.488968 http://dx. doi.org/10.1109/ICNN.1995.488968
    [14] EBERHART R C, SHI Y H. Particle swarm optimization: developments, applications and resources[C]//Proceedings of the 2001 Congress on Evolutionary Computation (IEEE Cat. No. 01TH8546). 2001. doi: 10.1109/CEC. 2001.934374 http://dx. doi.org/10.1109/CEC.2001.934374
    [15] SHI Y H, EBERHART R C. A modified particle swarm optimizer[C]//Proc of IEEE Icec Conference. 2009
    [16] CORTES C,VAPNIK V. Support-vector networks[J]. Machine Learning,1995,20(3):273-297. doi: 10.1007/BF00994018
    [17] HAYKIN S. 神经网络与机器学习[M]. 北京: 机械工业出版社, 2011.

    HAYKIN S. Neural networks and machine learning[M]. Beingjing: Neural Networks and Machine Learning, 2011.
    [18] BURGES C. A tutorial on support vector machines for pattern recognition[J]. Data Mining and Knowledge Discovery,1998,2(2):121-167. doi: 10.1023/A:1009715923555
    [19] ABOUHAWWASH M,SEADA H,DEB K. Towards faster convergence of evolutionary multi-criterion optimization algorithms using Karush Kuhn Tucker optimality based local search[J]. Computers & Operations Research,2017,79:331-346. doi: 10.1016/j.cor.2016.04.026
    [20] 韩玲. 基于人工神经网络: 多层感知器(MLP)的遥感影像分类模型[J]. 测绘通报,2004(9):29-30,42. doi: 10.3969/j.issn.0494-0911.2004.09.010

    HAN L. The classification model of RS images based on artificial neural network—MLP[J]. Bulletin of Surveying and Mapping,2004(9):29-30,42. doi: 10.3969/j.issn.0494-0911.2004.09.010
  • 加载中
图(11) / 表(3)
计量
  • 文章访问数:  1305
  • HTML全文浏览量:  102
  • PDF下载量:  69
  • 被引次数: 0
出版历程
  • 收稿日期:  2021-12-31
  • 修回日期:  2022-04-21
  • 录用日期:  2022-04-25
  • 刊出日期:  2022-10-01

目录

    /

    返回文章
    返回

    重要公告

    www.syltlx.com是《实验流体力学》期刊唯一官方网站,其他皆为仿冒。请注意识别。

    《实验流体力学》期刊不收取任何费用。如有组织或个人以我刊名义向作者、读者收取费用,皆为假冒。

    相关真实信息均印刷于《实验流体力学》纸刊。如有任何疑问,请先行致电编辑部咨询并确认,以避免损失。编辑部电话0816-2463376,2463374,2463373。

    请广大读者、作者相互转告,广为宣传!

    感谢大家对《实验流体力学》的支持与厚爱,欢迎继续关注我刊!


    《实验流体力学》编辑部

    2021年8月13日